Virginia Techii Synthesis, Characterization and Modification of Sulfonated Poly(arylene ether...

i

Synthesis, Characterization and Modification of Sulfonated Poly(arylene

ether sulfone)s for Membrane Separations

Ozma Norma Redd Pierce Lane

Dissertation submitted to the faculty of the Virginia Polytechnic Institute and State

University in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

in

Macromolecular Science and Engineering

Judy Riffle, Chair

S. Richard Turner

Richey Davis

Sue Mecham

Bruce Orler

(Blacksburg, VA)

Keywords: reverse osmosis, desalination, membrane electrolysis, fuel cells,

morphology, membrane fabrication, poly(arylene ether sulfone)

Copyright 2015 by Ozma Norma Redd Pierce Lane

ii

Synthesis, Characterization and Modification of Sulfonated Poly(arylene ether sulfone)s for

Membrane Separations

Ozma Norma Redd Pierce Lane

ABSTRACT

Sulfonated poly(arylene ether sulfone)s are a class of engineering thermoplastics well-

known for their mechanical properties and chemical/oxidative stability. The research in this

dissertation focuses on modifying the structure of sulfonated poly(arylene ether sulfone)s to

improve membrane performance. Blends of a 20% disulfonated poly(arylene ether sulfone)

(BPS20) with poly(ethylene glycol) (PEG) were investigated with the objective of promoting

water flux across a reverse osmosis membrane.

It was considered desirable to investigate poly(arylene ether sulfone)s with a

hydroquinone unit that could be controllably post-sulfonated without degradation, providing a

polymer with controlled sulfonation through controlling hydroquinone content. It also avoided

the disadvantages noted previously in polymers with post-sulfonated biphenol units. Initial

experiments focused on determining sulfonation conditions to confirm quantitative sulfonation of

the hydroquinone without side reactions or degradation. A polymer with 29 mole %

hydroquinone-containing units was used to study the rate of sulfonation. Successful post-

sulfonation was confirmed and reaction conditions were applied to a series of polymers with

varying hydroquinone comonomer contents. These polymers were sulfonated, characterized and

evaluated for transport properties. Of interest was the high sodium rejection in the presence of

iii

calcium, which in the directly copolymerized disulfonated materials is compromised. The post-

sulfonated poly(arylene ether sulfone)s showed no compromise in sodium rejection in a mixed-

feed of sodium chloride and calcium chloride.

In the membrane electrolysis of water, Nafion’s high permeability to hydrogen,

particularly above about 80°C, results in back-diffusion of hydrogen across the membrane. This

reduces efficiency, product purity, and long-term electrode stability. Hydrophilic-hydrophobic

multiblock copolymers based on disulfonated and non-sulfonated poly(arylene ether sulfone)

oligomers feature a lower gas permeability. Various multiblock compositions and casting

conditions were investigated and transport properties were characterized. A multiblock

poly(arylene ether sulfone) showed a significant improvement in performance over Nafion at

95°C.

Multiblock hydrophilic-hydrophobic poly(arylene ether sulfone)s have been extensively

investigated as alternatives for proton exchange membrane fuel cells. One concern with these

materials is the complicated multi-step synthesis and processing of oligomers, followed by

coupling to produce a multiblock copolymer. An streamlined synthetic process was successful

for synthesizing membranes with comparable morphologies and performance to a multiblock

synthesized via the traditional method.

iv

In memory of James E. McGrath

v

Acknowledgements

A PhD is an individual achievement made possible only by the contributions of the many

people surrounding that individual. I am very grateful for the support of friends, family,

colleagues, professors, and countless others who have helped me along my journey.

First and foremost, none of my work would have been possible without the support,

guidance, encouragement, and direction of Dr. James E. McGrath. His tireless dedication to his

research and his students is inspirational, and I will always be appreciative of the opportunity I

have had in studying with him. I also appreciate the herculean effort made by Dr. Judy Riffle in

taking on myself and my fellow graduate students from the McGrath group after his tragic illness

and passing. It was an extremely generous offer to make, and she has given us her own support,

encouragement, and direction. I am very fortunate to have had two truly exceptional advisers in

the course of my research, and I will never forget it.

I would also like to thank my committee members, Dr. Bruce Orler, Dr. Richard Turner,

Dr. Richey Davis, and Dr. Sue Jewel Mecham. In particular I would like to thank Sue for her

tireless support and direction during her time at Virginia Tech. I benefited from her insights into

my research and writing, and she deserves special recognition as well for her support and

direction of the McGrath group during its difficult times.

I would like to thank Steve McCartney and Chris Winkler at the Nanoscale

Characterization and Fabrication Laboratory of the Institute for Critical Technologies and

Applied Sciences (NCFL-ICTAS) for their training and assistance with the instrumentation there.

You made microscopy into the best part of my research, and you were excellent company while I

vi

investigated the many fascinating materials from the McGrath-Riffle groups. You will both be

missed.

I would not have been able to do any of my research without the support of the

administrative staff, who work tirelessly behind the scenes so that graduate students are

supported and materials can be ordered, and microscopes maintained. Laurie Good, Tammy Jo

Hiner, Cyndy Graham, and Angie Flynn are just a few of the many dedicated people who have

made my life easier at Virginia Tech. Thank you all.

Within the lab, I have had countless hours of helpful discussions and entertaining

conversations with my collaborators. In particular, I would like to thank the synthetic chemists

Hae-seung Lee, Ben Sundell, Jarrett Rowlett, Yu Chen, Natalie Arnett, Rachael VanHouten,

Amin Daryaei, Mou Paul, Ali Nebipasagil and Xiang Yu for the endless variety of new

copolymers which I had the pleasure of investigating both from morphological and performance

perspectives. When the time came for me to learn synthetic chemistry, Ben Sundell, Jarrett

Rowlett, and Amin Daryaei were generous with advice and feedback. For discussions on

microscopy, morphology, and structure-property-processing relationships, I would like to thank

Chang Hyun Lee, Myoungbae Lee, Hae-seung Lee, Anand Badami, and Abhishek Roy.

I have also been extremely fortunate in having an extended family to provide support and

many listening ears when I called, exasperated at some roadblock in my research, needing to be

heard and probably not making a lot of sense. To my Aunts Mary Jo, Stacy, Becky, Shelly,

Miriam, and Mary, thank you for the reminder that even Darwin had very bad days. Thank you

for being there, and for listening, and for the care packages. To my uncles David, Andy, Paul,

Larry, and Joe, you have been there for advice (solicited or not), for support, and for incredible

vii

patience. Thank you to my countless cousins, who not only ask about my research but listen

politely when I start talking about it. I am grateful to my siblings Seana and Lex, and my niblings

Lane and Eleanor. You are the best and I love you all. And last but certainly not least, I would

like to thank my mom, Caroline Lane.

viii

ATTRIBUTION

Chapter 2.

Chang Hyun Lee, Ph.D. (Department of Macromolecular Science and Engineering) was a post-

doctoral research at Virginia Tech. Dr. Lee was a co-author on this paper and prepared the blends

investigated in the research.

James E. McGrath, Ph.D. (Department of Macromolecular Science and Engineering) was a

professor at Virginia Tech. Dr. McGrath was a co-author on this paper, co-principal investigator

for the grant supporting this research and contributed revisions to the final document.

Jianbo Hou, Ph.D (Deparment of Chemistry) was a post-doctoral researcher at Virginia Tech. Dr.

Hou contributed NMR analysis and revisions to the final document.

Louis A. Madsen, Ph.D (Department of Chemistry) is a professor at Virginia Tech. Dr. Madsen

is a co-author on this paper, co-principal investigator for the grant supporting this research and

contributed analysis and revisions to the final document.

Justin Spano, Ph.D (Department of Chemistry) was a graduate researcher at Virginia Tech. He

performed thermal analysis and contributed revisions to the final document.

Sungsool Wi, Ph.D (Department of Chemistry) was a post-doctoral researcher at Virginia Tech.

He perfomed NMR analysis and contributed revisions to the final document.

Joseph R. Cook, Ph.D. (Department of Chemical Engineering) was a graduate researcher at the

University of Texas at Austin. Dr. Cook was a co-author on this paper and contributed to the

transport property testing of the prepared blends.

ix

Wei Xie, Ph.D (Department of Chemical Engineering) was a graduate researcher at the

University of Texas at Austin. Dr. Xie was a co-author on this paper and contributed to the

transport property testing of the prepared blends.

Geoffrey Geise, Ph.D (Department of Chemical Engineering) was a graduate researcher at the

University of Texas at Austin. Dr. Geise was a co-author on this paper and contributed to the

modeling and analysis of blend processing conditions and their impact on transport properties.

Benny D. Freeman, Ph.D. (Department of Chemical Engineering) is currently a professor at the

University of Texas at Austin. Dr. Freeman was a co-author on this paper and contributed

revisions to the final document.

Chapter 3. Kinetics of Post-Sulfonation of a Poly(arylene ether sulfone) that Contains

Hydroquinone (Radel-A) for Application in Reverse Osmosis Membranes

Sue J. Mecham, Ph.D. (Department of Macromolecular Science and Engineering) is currently a

research scientist at the University of North Carolina at Chapel Hill. Dr. Mecham was a co- author

on this paper and contributed to the SEC analysis of the polymers.

Eui-Soung Jang, B.S. (Department of Chemical Engineering) is currently a graduate student at

the University of Texas at Austin. Mr. Jang was a co-author on this paper and contributed to the

water purification property testing of the prepared films.




x

Judy S. Riffle, Ph.D. (Department of Macromolecular Science and Engineering) is a professor at

Virginia Tech. Dr. Riffle was a co-author on this paper and contributed revisions to the final

document.



for the grant supporting this research.

Chapter 4. Synthesis, Characterization and Post-sulfonation of a Polysulfone Series

Incorporating Hydroquinone for Reverse Osmosis Membranes.

Eui-Soung Jang, B.S. (Department of Chemical Engineering) is currently a graduate student at

the University of Texas at Austin. Mr. Jang is a co-author on this paper and contributed to the

water purification property testing of the prepared films.

Shreya Roy Choudhury, B.S. (Department of Chemistry) is currently a graduate student at

Virginia Tech. Ms. Choudhury is a co-author on this paper and performed the molecular

weight characterization of the polymers.


research scientist at the University of North Carolina at Chapel Hill. Dr. Mecham was a co- author

on this paper and contributed to the SEC analysis of the polymers.



xi




document.


professor at Virginia Tech. Dr. McGrath was a co-author on this paper and co-principal

investigator for the grant supporting this research.

Chapter 5. Synthesis and Characterization of a Hydrophilic-Hydrophobic Multiblock

Copolymer for Membrane Electrolysis Applications

Jarrett Rowlett, Ph.D (Department of Macromolecular Science and Engineering) was a graduate

researcher at Virginia Tech. Dr. Rowlett was a co-author on this paper and synthesized the

multiblock copolymers studied in this research.

Cortney Mittlesteadt, Ph.D. is a researcher at Giner. Dr. Mittlesteadt is a co-author on this paper

and performed the electrolysis characterization of the membranes.

Jason Willey, Ph.D. is a researcher at Giner. Dr. Willey is a co-author on this paper and

performed the electrolysis characterization of the membranes

Amin Daryaei, B.S. (Department of Macromolecular Science and Engineering) is a graduate

researcher at Virginia Tech. Mr. Daryaei assisted with the multiblock polymer synthesis and is a

co-author on this paper.

xii






document.


research scientist at the University of North Carolina at Chapel Hill. Dr. Mecham was a co-

author on this paper and contributed to the SEC analysis of the polymers.

Chapter 6. Synthesis and Characterization of Hydrophobic-Hydrophilic Block Copolymers for

Proton Exchange Membranes

Rachael A. VanHouten, Ph.D was a graduate researcher at Virginia Tech. Dr. Vanhouten is a co-

author on this paper and performed the synthesis of the materials studied therein.

Desmond J. VanHouten, Ph.D is currently a research scientist at Corning. Dr. Vanhouten is a co-

author on this paper and performed thermomechanical analysis on the materials studied in this

paper.






xiii

document.

xiv

Table of Contents

1.0 Literature Review ............................................................................................................ 1

1.1 The Need for Development of Clean Water Production................................................ 1

1.2 Osmotic Pressure and Reverse Osmosis .......................................................................... 4

1.3 Thermal Processes of Desalination .................................................................................. 6

1.3.1 Vapor Compression ................................................................................................... 7

1.3.2 Multi-stage Flash Distillation and Multi-effect Distillation ............................... 7

1.3.3 Membrane Distillation .............................................................................................. 8

1.4 Overview of Membrane Processes of Desalination and Other Separations ...................... 9

1.4.1 Microfiltration and Ultrafiltration ....................................................................... 10

1.4.2 Nanofiltration ........................................................................................................... 13

1.4.3 Reverse Osmosis ...................................................................................................... 14

1.4.4 Electrodialysis/Electrodialysis Reversal ............................................................ 17

1.4.5 Cost effectiveness and comparative advantages of different desalination processes ............................................................................................................................. 18

1.4.6 Pressure Retarded Osmosis and Forward Osmosis .......................................... 21

1.4.7 Pervaporation........................................................................................................... 29

1.5 Design of the Reverse Osmosis Process ..................................................................... 30

1.5.1 Solution-Diffusion Mechanism .............................................................................. 30

1.5.2. Terms and Equations in Membrane Transport Theory .................................. 34

1.5.3 Pre-treatment ........................................................................................................... 36

1.5.4 Membrane treatment .............................................................................................. 38

1.5.5 Post-treatment ......................................................................................................... 41

1.6 Challenges in reverse osmosis ..................................................................................... 41

1.6.1 Fouling........................................................................................................................ 42

1.6.2 Concentration Polarization.................................................................................... 48

1.6.3 Removal of heavy metals ........................................................................................ 49

1.6.4 Removal of boron ..................................................................................................... 50

1.6.5 Sustainable power use ............................................................................................ 50

1.6.6 Brine Disposal .......................................................................................................... 52

1.7 Membranes and composites used in reverse osmosis and osmotic power ........ 53

xv

1.7.1 Materials used in reverse osmosis ....................................................................... 53

1.7.2 Materials used in pressure retarded osmosis/forward osmosis ................... 56

1.7.3 Asymmetric membranes and Thin Film Composites........................................ 57

1.7.4 Phase inversion ........................................................................................................ 59

1.7.5 ASM finishing techniques ....................................................................................... 65

1.8 Sulfonated poly(arylene ether sulfone)s and membrane separations ................ 66

1.8.1 Sulfonated poly(arylene ether sulfone)s ................................................................... 66

1.8.2 Post-sulfonated poly(arylene ether sulfone)s and reverse osmosis ............. 67

1.9 Characterization of RO Membranes ............................................................................ 70

1.9.1 Mechanical properties ............................................................................................ 71

1.9.2 Thermomechanical behavior via dynamic mechanical analysis ................... 73

1.9.3 Scanning Electron Microscopy .............................................................................. 74

1.9.4 Atomic Force Microscopy ....................................................................................... 75

2.0 Disulfonated Poly(arylene ether sulfone) Random Copolymer Blends Tuned for

Rapid Water Permeation via Cation Complexation with Poly(ethylene glycol) Oligomers

................................................................................................................................................... 88

2.1 Introduction ................................................................................................................... 89

2.2 Experimental .................................................................................................................. 92

2.3 Results and Discussion .................................................................................................. 96

2.4 Conclusions................................................................................................................... 111

3.0 Kinetics of Post-Sulfonation of a Poly(arylene ether sulfone) that Contains Hydroquione

(Radel-A) for Application in Reverse Osmosis Membranes ................................................. 117

3.1 Introduction ..................................................................................................................... 117

3.2 Experimental ....................................................................................................................... 120

3.2.1 Materials ....................................................................................................................... 120

3.2.2 Sulfonation of Radel A ................................................................................................. 120

3.2.3. 1H NMR..................................................................................................................... 121

3.2.4. Water Uptake............................................................................................................... 121

3.2.5 Size Exclusion Chromatography................................................................................. 121

3.3 Results and Discussion ........................................................................................................ 122

3.3.1. Post-sulfonation of Radel A ........................................................................................ 122

3.3.2 Kinetics ...................................................................................................................... 127

3.3.2. Molecular weights ....................................................................................................... 130

3.4 Conclusions .......................................................................................................................... 131

xvi

4.0 Synthesis, Characterization and Post-sulfonation of a Polysulfone Series Incorporating

Hydroquinone for Reverse Osmosis Membranes .................................................................. 133

4.1 Introduction ......................................................................................................................... 133

4.2 Experimental ....................................................................................................................... 134

4.2.1 Materials ....................................................................................................................... 134

4.2.2. Synthesis and Sulfonation ........................................................................................... 135

4.2.3. Characterization .......................................................................................................... 136

4.2.3.1 NMR ........................................................................................................................ 136

4.2.3.2. Water Uptake ........................................................................................................ 136

4.2.3.3. Molecular Weight Characterization ................................................................... 136

4.2.3.4 Transport Property Measurement ....................................................................... 137

4.3 Results and Discussion ........................................................................................................ 138

4.3.1 Synthesis ........................................................................................................................ 138

4.3.2 Membrane Properties .................................................................................................. 139

4.3.3 Transport Data ............................................................................................................. 140

4.4 Conclusions .......................................................................................................................... 142

5.0 Synthesis and Characterization of a Hydrophilic-Hydrophobic Multiblock Copolymer

for Membrane Electrolysis Applications ................................................................................ 144

5.1 Introduction ..................................................................................................................... 145

5.2 Experimental ....................................................................................................................... 146

5.2.1 Materials ....................................................................................................................... 146

5.2.3 Synthesis ........................................................................................................................ 147

5.2.3.1 Synthesis of a HQSH100 Hydrophilic Block ....................................................... 147

5.2.3.2. Synthesis and Endcapping of Hydrophobic Blocks ........................................... 148

5.2.3.3 Coupling Reaction of the Hydrophilic and Hydrophobic Blocks...................... 148

5.2.4.1 Synthesis of the Fluorine-Terminated Hydrophobic Block ............................... 149

5.2.4.2 Coupling of the Hydrophilic and Fluorine-Terminated Hydrophobic Blocks 149

5.2.5 Characterization ........................................................................................................... 150

5.2.5.1 NMR spectroscopy ................................................................................................. 150

5.2.5.2 Film Casting, Annealing, and Acidification ........................................................ 150

5.2.3.3 Water Uptake ......................................................................................................... 150

5.2.3.3 Transmission Electron Microscopy ..................................................................... 151

5.3 Results .................................................................................................................................. 151

5.3.1 Synthesis ........................................................................................................................ 151

5.3.1.1 Hydrophilic Oligomer Synthesis .......................................................................... 151

xvii

5.3.1.2 Synthesis of the Hydrophobic Oligomer .............................................................. 153

5.3.1.3 Coupling of Hydrophilic and Hydrophobic Oligomers...................................... 156

5.3.1.4 Bulk Morphology ................................................................................................... 159

5.3.1.5 Effect of Annealing and Molecular Weight ......................................................... 160

5.3.1.6 Impact of Casting Solution on Film Morphology ............................................... 161

5.3.1.7. Conductivity and Permeability............................................................................ 162

5.4 Conclusions .......................................................................................................................... 163

6.0 Synthesis and Characterization of Hydrophobic-Hydrophilic Block Copolymers for

Proton Exchange Membranes .................................................................................................. 166

Abstract ...................................................................................................................................... 166

6.1 Introduction ......................................................................................................................... 167

6.2 Experimental Section .......................................................................................................... 170

6.2.1 Materials ....................................................................................................................... 170

6.2.2. Synthesis of Phenoxide-Terminated Hydrophilic Blocks (BPS-100). ..................... 171

6.2.3. Synthesis of Segmented Copolymers. ........................................................................ 171

6.2.4. Synthesis of BisSF-BPSH100 Multiblock Copolymer Controls. ............................. 172

6.2.4.1. Synthesis of Fluorine-Terminated Hydrophobic Blocks (BisSF) (2). .............. 172

6.2.4.2. Synthesis of Phenoxide-Terminated Hydrophilic Blocks (BPS-100) (1).......... 173

6.2.4.3. Synthesis of BisSF-BPS100 Multiblock Copolymers. ........................................ 173

6.2.5. Characterization of Copolymers. ............................................................................... 174

6.2.6. Membrane Preparation. ............................................................................................. 174

6.2.7. Determination of Water Uptake and Dimensional Swelling. .................................. 174

6.2.8. Measurement of Proton Conductivity. ...................................................................... 175

6.2.9. Tensile Testing. ............................................................................................................ 176

6.2.10. Surface Morphology. ................................................................................................. 176

6.2.11. Bulk Morphology....................................................................................................... 176

6.3. Results and Discussion ....................................................................................................... 177

6.3.1. Synthesis of Hydrophilic Oligomers. ......................................................................... 177

6.3.2. Synthesis of BisSF-BPSH100 Segmented Copolymers. ........................................... 180

6.3.3. Synthesis of BisSF-BPSH100 Multiblock Copolymer Controls. ............................. 182

6.3.4. Comparison of BisSF-BPSH100 Segmented and Multiblock Copolymer Properties.

................................................................................................................................................. 184

6.4. Conclusions ......................................................................................................................... 190

7.0 Future Work ........................................................................................................................ 193

7.1 Post-sulfonated hydroquinone-containing poly(arylene ether sulfone)s ................ 193

xviii

7.2 Structure-property-processing relationships in multiblock copolymers ................ 195

1

1.0 Literature Review

Drinkable clean water has been widely available in the United States for most of recent

history. However, increasingly parts of the United States as well as large regions in the rest of the

world have either become stressed for sufficient supplies of drinkable water, or are dealing with

absolute shortages of even non-potable water. Over 1 billion people have no access to safe, clean

drinking water and over 2 billion people experience regular water shortages. Regions which may

be only slightly stressed for water at present are projected to suffer increasingly severe water stress

and shortages in the future as a result of increasing population, shifting patterns of water

distribution due to climate change, and decreasing supplies of drinkable fresh water.1, 2 As fresh

water supplies become more stressed and contamination of remaining supplies more severe,

desalination of ocean water and brackish water is the most obvious solution.

In anticipation of the need to produce greater amounts of clean water in the future, there is

a need both for more water production facilities, and for an improvement in efficiency of the

systems currently in use.3 A number of different water treatment methods exist which produce

water, either with thermal phase changes (primarily distillation) or using a membrane separation

process. While multistage flash (MSF) distillation was originally the most economical and most

reliable method of desalination, the rapid development of membrane separation technology in the

1970s and 1980s has resulted in more competitive technology that surpassed MSF in production.4

The most affordable and efficient commercial method for producing clean drinking water from

seawater and brackish water is reverse osmosis.3, 5

1.1 The Need for Development of Clean Water Production

2

The United States Geological Survey estimates that less than one percent of the available

water on the planet is viable, available freshwater, with the remainder ranging from slightly

brackish to saltwater, with some additionally tied up in the polar ice caps.4 Given the present

scarcity of clean drinking water and the increasing need for the future due to the pressures of

population growth, agriculture, industry, and a decreasing supply due to pollution, there is an

urgent need to develop economically advantageous solutions for producing larger volumes of

clean, affordable drinking water.3, 6 As water treatment and water management have been

recognized as increasingly critical issues in recent decades, projections of future water demands

continually stress the need for development of newer, more efficient means of providing potable

water.7-9 Further complicating the issue is the requirement of available, affordable energy for clean

water production, regardless of whether it is desalination or water recycling. Areas experiencing

or projected to experience water stress frequently also lack the infrastructure for supplying reliable

and affordable energy which can be dedicated to drinking water production.3, 10, 11 Combined, these

factors point to a future where water supply concerns meet or exceed those of fossil fuel

availability, and research to help supply those needs before they become critical is essential.4, 12-14

In addition to a lag in planning and preparation for meeting future water demands, shifts in

water supply or reliability of water supply are also projected due to changes in precipitation

patterns as a result of potential climate change. While areas of the world outside the United States

have been experiencing water stress or water shortages for some time and have responded with an

exponential increase in desalination capacity, the rate of increase in desalination capacity in the

United States has been less dramatic.7 However, there are now desalination plants or membrane-

based water treatment plants in almost every state, with over 75% of production directed towards

municipal or industrial supplies. 7 Across the country, desalination plants are most commonly

3

found in California, and reverse osmosis plants have increased sharply both in total number and in

relative share of the water treatment methods since 1995.13, 15, 16

While there are different water filtration mechanisms to remove contaminants of varying

size, reverse osmosis is the most viable process for the removal of small ions such as those found

in seawater. Salts and suspended materials are removed from brackish water (2,000 to 5,000 ppm

salinity) and saltwater (35,000-45,000 ppm salinity) by desalination, and there are a variety of

desalination methods which will be reviewed.6 Acceptable total dissolved solids (TDS) limits for

potable water vary widely for different applications. The TDS limit for water classified as fresh

water is 1,000 ppm.4 Above this point the taste, color, odor, and corrosiveness can be affected.

There are also further limits on tolerable levels of individual molecular species for the consumption

of potable water.

There are frequently additional standards for water used for irrigation, which may vary by

country or state, such as the total allowed limit of chlorine, boron, or other components in addition

to the total dissolved solids (TDS) content.4 Of the desalination processes available, municipal

consumption uses a far greater percentage of treated water than industrial, military, agricultural,

power production, or other uses.3 There is also a growing need for the production of ultrapure

water used in the manufacturing of electronics and pharmaceuticals, both of which require a much

higher standard of purity than the production of drinking water.16-18 Furthermore, softened water

is required for use in boilers, and while this niche was previously served by ion exchange resins,

reverse osmosis is replacing or complementing this method by improving performance and

avoiding the maintenance costs of regenerating the ion exchange resin.16 The reverse osmosis

system is an ideal candidate to meet the needs of all of these markets, and will likely become more

4

effective with future improvements in performance and with greater demand for potable water.14,

16, 19, 20

While reverse osmosis (RO) is primarily used to treat seawater and brackish water, there

is an obvious economic advantage in also using RO to recover drinking water from domestic or

industrial wastewater. This water is generally more dilute than brackish water, and therefore it is

energetically cheaper than brackish water or seawater to separate, but it is too contaminated with

different components to be applicable for direct consumption. The membrane treatment of this

water therefore requires several steps for the separation of components of varying sizes, to be

discussed in further detail later in this chapter. The ionic exclusion component of the separation

process may still be able to tolerate lower levels of rejection from a cheaper, less efficient method

such as nanofiltration, as the overall requirements for TDS reduction are less severe than in the

case of seawater or brackish water. This is a newer focus of research and development as awareness

of water stress increases around the world, and is a potentially novel application for materials

which may be selective for a wider range of solutes or resistant to other pollutants.3-5 The steady

increase in demand for the production of drinkable water requires additional improvements in

efficiency and selection.4, 13, 21 There is also growing interest in membrane processes to remove

contaminants to produce or recycle water for agricultural use, and this requires less stringent

standards than for potable water production.22

1.2 Osmotic Pressure and Reverse Osmosis

When a selectively permeable membrane (one highly permeable to one component but

resistant to another) is used to separate solutions of different concentrations, the result of random

movement will be for the more permeable component to migrate across to the region of less

concentrated solution. This phenomenon is called osmosis, and the amount of pressure that must

5

be exerted on a given solution to prevent the influx of water from pure water is known as the

osmotic pressure. When the applied pressure to a given solution exceeds the osmotic pressure, the

direction of water flow reverses and pure water can be extracted from a pressurized solution.

The process of osmosis was first observed in 1748 by Abbe Nollet, but the relationship

between osmotic pressure and solute concentration was not mathematically described until the 19th

century by Jacobus Henrius Van’t Hoff. The relationship was improved shortly afterwards by

Harmon Northrup Morse to achieve the osmotic pressure equation used today.16 The osmotic

pressure, π, may be determined from the following Van’t Hoff equation:

𝜋 = 𝑖𝑀𝑅𝑇

Above, i is the unitless Van’t Hoff factor equal to the number of ions created by each atom

of dissolved solute, M is the solution molarity (mole/L), R is the gas constant (8.314 J/K*mole),

and T is the absolute temperature in degrees Kelvin. The Van’t Hoff equation calculates the

pressure created by a concentrated solution against a neat solvent, but the overall π between a

concentrated and dilute solution can be calculated from the difference of the individual osmotic

pressures. The osmotic pressure of a concentrated solution is defined as the pressure required to

resist the influx of pure water (as will be discussed later in the case of forward osmosis). In reverse

osmosis a higher pressure is applied in order to drive the migration of water in the opposite

direction, from the more concentrated to the less concentrated solution.4, 14, 15

While reverse osmosis is one of the primary mechanisms used for desalination, there are

other methods for separating relatively pure water from brackish or salt water, or for separating

dissolved or suspended components which are larger than monovalent salts. These methods may

6

be separated into two primary categories based on the mechanism of action: thermal separation

(phase change) processes and membrane (single phase) processes. 4, 14, 15

1.3 Thermal Processes of Desalination

Distillation is the oldest method of desalination, having been used for hundreds of years

before the advent of more refined processes, and it is still used today on an industrial scale.

Thermal processes (frequently using some application of distillation) are therefore better

established and are used more widely than the more recent membrane separation technologies.

However, they are losing market share due to their greater energetic and economic costs. As of

2008, thermal processes covered 43% of the global desalination market.7, 14, 22 Thermal processes

consist primarily of vapor compression and multi-effect flash distillation, although the majority of

the world’s thermal distillation plants now use multi-stage distillation. Thermal processes are

defined by the requirement of a phase change, although membrane distillation also incorporates a

semi-permeable membrane. Due to the phase change, thermal processes generally have the

advantage of simplicity over membrane separation processes, as high purity can be reached

without lengthy pre-treatment processes or extensive monitoring of feedwater components.

However, recovery levels range from 15-45%, which is generally lower than the ranges offered by

membrane processes such as reverse osmosis or nanofiltration.23-25

One of the primary concerns in a distillation system is conservation of the thermal energy

used to produce the phase changes. Energy conservation designs are typically centered on re-using

the heat from condensation to warm feedwater for the evaporation phase.7, 26 A new focus in

developing potable water production facilities is hybridization between reverse osmosis and

thermal distillation systems to optimize process design.15, 27

7

1.3.1 Vapor Compression

Vapor compression, also referred to as the mechanical vapor compression process is a

simple but low-output method of producing fresh drinking water from a saline feed source. It

operates by drawing vapor from the air above the feedwater/brine source and feeding it into a vapor

compressor. The compressed vapor, which is warmer than the feedwater due to compression, is

then introduced on the outside of the tubes through which the relatively cooler feedwater or brine

is transported, thus condensing the vapor while heating the feedwater. It may be the more favorable

of thermal desalination methods in an area with relatively cheap energy, but not in locations where

low-cost steam and cooling water are difficult to supply.7 This method is used primarily for

maintaining small amounts of drinkable water, generally less than 100 m3/day, at resorts and for

industrial purposes.7, 13, 28

1.3.2 Multi-stage Flash Distillation and Multi-effect Distillation

Multi-stage flash distillation (MSF) is the simplest distillation procedure used industrially,

and has been in common use since the 1950s. It is more frequently found in the Middle East, where

water supplies are stressed and the cost of energy is comparatively low. Its relative simplicity and

reliability is an advantage over membrane separation processes.23, 25 Significant advantages are

that the feedwater has a much lower pretreatment requirement, and the product water features a

higher output than RO, and also a much lower level of total dissolved solids.7, 23, 29

The MSF process entails passing seawater through a series of chambers where it is boiled

and the water vapor is collected as the pressure becomes progressively lower. By reaching the end

of the series of chambers, the remaining seawater has been concentrated into heated brine. This is

then pumped back through the tanks in order to re-use the heat for evaporating feedwater

downstream as well as to conserve costly water-conditioning chemicals.29 The brine is corrosive

8

and in addition to the expense of thermal energy, MSF requires labor to inspect and frequently

replace fittings used in the chambers. The process also requires a large amount of space for the

distillation chambers.23, 29 A major disadvantage is the steep energetic cost of producing water at

18 kWh/m3 which is much higher than the 5 kWh/m3 of reverse osmosis.13, 23, 30

Multi-effect distillation (MED) is a variation on the MSF process in two aspects. First, the

pipes used for vaporization and condensation tend to be horizontal rather than vertical. Secondly

and more importantly, MED features a more efficient use of heat than MSF at 15 kWh/m3.

Although it has the advantage of greater energetic efficiency and was developed prior to MSF, it

features a lower output and has not been as widely utilized.23, 29 MED was commercially

implemented prior to MSF, and the first commercial MED plant was established in 1928 in the

Netherlands.13, 23

1.3.3 Membrane Distillation

Membrane distillation is a process which is thermally driven. It incorporates a

microporous hydrophobic membrane that is impermeable to liquid water but will allow water

vapor to pass through the pores.31 A typical membrane distillation unit contains a high-

temperature flow of feedwater on one side of a condensation chamber containing a microporous

membrane, producing high levels of water vapor. The water vapor diffuses across the

membrane into the condensation chamber, where it cools and condenses on a plate maintained

at a lower temperature. The coolant used in the condensation plate is then cycled back to use the

newly absorbed heat to raise the temperature of the feed water.7 As with other thermal

processes, membrane distillation is energetically expensive and could be more viable in a

system where some or all of the energetic expense can be provided by alternative energy.31, 32

Compared to other distillation systems, it is not yet competitive and is primarily in the stage of

9

research and development.3, 30, 33-36

1.4 Overview of Membrane Processes of Desalination and Other

Separations

Membrane separation processes, referred to as single-phase processes, entail different

degrees of separation and can most easily be categorized by the size of the particles or molecules

which are being excluded and the type of force driving the separation. Typically, this force is

primarily applied pressure but may also include species concentration, temperature and voltage.29

The discussion here is confined to processes used in the production of potable water and focuses

on the removal of ions and particles from water. Similar principles, however, may also apply to

separations of gases and vapors. The five common processes used in the treatment of water for

human consumption are reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF),

microfiltration (MF), and electrodialysis/electrodialysis reversal (ED/EDR).14 The separation

characteristics of the first four methods are illustrated in Figure 1-1. The exclusion process takes

place either via contaminants being physically too large to enter the pores of a membrane, in the

case of size exclusion, or by the components being unable to diffuse into the membrane and across

through the free volume as the permeate, in the case of solution-diffusion.

These processes are frequently used in combination, as they may be used to remove

contaminants of differing sizes and functionalities. Microfiltration and ultrafiltration in particular

are very similar with respect to operating conditions and materials, varying primarily in the size of

excluded particles. Of these five processes, only nanofiltration, reverse osmosis and

electrodialysis/electrodialysis reversal can be used in desalinating water, and reverse osmosis is

the most efficient method.4 However, the other methods are described as they are frequently used

10

in conjunction with reverse osmosis or nanofiltration as pre-treatment processes. While these

processes are primarily used to desalinate or decontaminate water, they may also be used to

concentrate dissolved or suspended solids of interest in the feedwater. This process is termed

dewatering and is used in certain industrial and agricultural applications.16 Of those three

processes, reverse osmosis is the only practical process for desalination of seawater. The other two

methods are more suited for brackish conditions, remediation of contaminated freshwater, or

removal of particulates for agricultural or industrial applications which do not require complete

purification of feed water.4, 37 Additionally, membrane processes not used for the production of

drinkable water such as forward osmosis, pressure retarded osmosis, and pervaporation will be

discussed as they feature very similar concepts and use similar or identical materials to those used

in reverse osmosis.15, 16

Figure 1-1. Comparison of pore size and exclusion capacity of membrane separation systems.19, 26

1.4.1 Microfiltration and Ultrafiltration

Microfiltration and ultrafiltration are pressure-driven processes that use a sieving

mechanism through pores in a membrane to screen out particles above a given size. Due to the

11

exclusion mechanism, the primary disadvantage to both of these systems is the need to manage

membrane fouling. Both MF and UF have applications in food, agriculture, and industry for

concentrating products and removing water or unwanted components. They also are used in the

pretreatment phase of reverse osmosis for removing particles or microbes which might cause

fouling of the RO membrane.38, 39 The first large-scale MF/UF plant in the USA began operations

in 1994, and so this is still a relatively new field with opportunities for research and development.

As costs have dropped rapidly due to economies of scale and technological innovation, the number

of MF or UF installations has increased significantly.5, 40

The primary advantages of MF and UF over conventional water treatment processes which

do not require a membrane are 1) the filtration mechanism allows for selective removal of

particulates larger than the given pore size without the expense of physicochemical removal, and

2) the highly uniform pore sizes afforded by MF and UF membrane fabrication techniques allow

for nearly quantitative elimination of the target solids or microbes. Levels of a given particulate or

microorganism can be reduced by several orders of magnitude and so reductions are described in

‘logs’ of reduction, rather than the percent rejection used in other membrane treatment processes.40

Ultrafiltration is a pressure-driven process that can remove bacteria and protozoa from the

water. It is typically intended to screen out viruses and large organic macromolecules on the order

of 5,000-100,000 g/mole. Ultrafiltration cannot reliably screen out smaller macromolecules or

dissolved species, and both internal and external fouling are significant challenges. The membrane

is typically comprised of cellulose acetate, polypropylene, or other synthetic polymers, which are

frequently formed into hollow fibers but may include membrane sheets in either spiral-wrapped,

tubular, or plate-and-frame configurations.5 Hollow fibers predominate due to unparalleled

packing density, high surface area and low energy consumption. Regardless of module geometry,

12

ultrafiltration membranes tend to be anisotropic structures formed by phase inversion.5, 14

Ultrafiltration can also be useful for separation and/or concentration of agricultural and dairy

products.37, 40

Microfiltration (MF) is a pressure-driven process wherein microporous membrane sieves

are used to remove particulates, precipitates, and very small microorganisms (such as Giardia

lamblia and Cryptosporidium parvum, both pathogens of concern for immunocompromised

individuals) from feedwater.14 It cannot remove dissolved species, or particles smaller than about

100 nm or under about 100,000 g/mole. It is frequently utilized as a pre-treatment step for

removing compounds which are likely to cause fouling from the feedwater during a reverse

osmosis process. Microfiltration is also a valuable component in separations in the food and

agricultural industry. Like ultrafiltration, operations must contend with internal and external

fouling, which require constant surveillance and treatment.37, 40

Microfiltration frequently uses hollow fibers or membranes made from cellulose acetate,

polypropylene, polysulfone, poly(vinylidene fluoride), poly(ether sulfone), or other materials.14

An important feature of MF membranes, as with ultrafiltration membranes, is the uniformity of

the pore size, assuring ‘absolute’ or near-complete removal of a given particle or microbe which

is more difficult to accomplish when using coagulation-based water treatment membranes.40 The

sieving mechanism of microfiltration has a principal disadvantage in the necessity for managing

fouling, which is the focus of a significant portion of MF research.40-42

Hollow fibers have rapidly become a standard geometry for microfiltration due to efficient

production via a superior surface area to volume ratio per unit of module space, as well as bi-

13

directional strength which allows for designs incorporating either inside-out or outside-in flow and

backwashing with air and/or water.5, 40

1.4.2 Nanofiltration

Nanofiltration (NF) is a pressure-driven process which can be used to treat ‘hard’ water by

removing divalent ions, and may partially remove certain monovalent ions as well as dissolved

organic molecules over approximately 300 g/mole. It can filter out larger particles and single-

celled organisms as well, but this carries a higher risk of fouling. Sometimes nanofiltration is used

to refer to reverse osmosis, but the processes differ, as NF uses both solution-diffusion and

size/charge exclusions in the membrane separation process.7, 14 While nanofiltration membranes

may feature a dense layer, they also frequently feature pores ranging from 1-10 nm in diameter,

which will permit passage of most monovalent salts. The required pressure in feedwater to produce

permeate in nanofiltration is approximately 70-120 psi, an order of magnitude lower than pressures

required in reverse osmosis. The primary advantage of nanofiltration over reverse osmosis is that

of cost due to the lower energy and pressure requirements, but the limitations of NF constrain it to

niche applications of water purification.14, 16

Nanofiltration was developed in the 70s and refined in the 80s as a variation of reverse

osmosis membranes that required lower operating pressures and had lower rejection mandates.4

As in reverse osmosis, nanofiltration membranes are typically either cellulose acetate or

crosslinked polyamides, and the majority of modules use the spiral-wound configuration

(explained in detail in Section 1.5.4).14 While nanofiltration is unable to adequately desalinate

feedwater with a high salt concentration, such as sea water or the highly saline waters found in the

Middle East, it can desalinate brackish water and performs well in applications such as water

softening. It may have an additional niche in boron removal.4, 14, 43-45 It can also be used as a lower-

14

cost alternative to reverse osmosis for concentrating agriculture and dairy products.37 There is a

growing interest in using nanofiltration to remediate produced or processed water which has

become contaminated in the process of fossil fuel extraction.46, 47 Like reverse osmosis,

nanofiltration uses a multistep process of pre-treatment, membrane treatment, and post-treatment

of water prior to distribution for consumption.2, 5, 14

1.4.3 Reverse Osmosis

Reverse osmosis (RO) is a pressure-driven process in which ions are removed from

solution via selectively permeable membranes such as cellulose acetate, polyamides, or sulfonated

poly(arylene ether sulfone)s. These materials are different from those seen in microfiltration and

ultrafiltration membranes because they require water to be absorbed into and diffused across the

membrane, necessitating a degree of hydrophilicity. Although RO is capable of separating larger

molecules and particles, in practical applications these must be reduced or eliminated in an earlier

treatment step (using microfiltration, ultrafiltration, or a coagulating/flocculating technique and

possibly nanofiltration) in order to minimize fouling, and additional fouling prevention methods

are sometimes necessary.39, 48

Reverse osmosis requires 3-10 kWh per cubic meter of freshwater produced, which makes

it one of the most energetically affordable desalination technologies.1, 31 The specific energy

requirements for any given RO system are determined by the salinity of the water to be treated,

and the properties of the selected RO membrane. An ideal RO membrane has excellent water

transport with near-complete salt rejection, a very thin, pinhole-free selective skin layer for

promoting rapid water transport, and must have sufficient mechanical properties to withstand the

driving pressure of the process.3, 15 In order to improve flux, a free-standing dense membrane is

unsuitable and either a thin film composite (a submicron layer of selective material supported by

15

a layer of polysulfone foam and a nonwoven fabric to provide mechanical integrity) or an

asymmetric membrane (a thin film and support foam cast simultaneously from the same material)

are used to optimize membrane performance.14 These structures and their fabrication will be

discussed at a later section in much greater detail.

The first experimental membranes developed for desalination were dense cellulose acetate

membranes made by Reid and Breton in 1959, after evaluating a series of different partially

hydrophilic materials for permeability of water and for salt rejection.49 While they established

satisfactory rejection values, the water flux was too low. Loeb and Sourirajan developed

asymmetric membranes from cellulose acetate in the 1960’s to improve water flux while

maintaining satisfactory salt rejection values. This resulted in the commercial viability of RO

membranes and the birth of the RO industry.14, 49 Improvements in the fabrication and design of

membranes led to a rapid growth in the performance of commercial RO units in the late 1960’s

and early 1970’s, the most significant being the development of spiral-wound and then tubular or

hollow fiber membranes to replace flat-sheet membranes. The introduction of linear aromatic

polyamides provided a much higher water flux at a slightly lower, and therefore more affordable,

pressure.16 Another breakthrough with an interfacially-produced crosslinked polyamide was made

by John Cadotte, in the form of a thin film composite. This offered advantages in both water flux

and salt rejection, and has since been established by Dow Chemical Company as the predominant

industry standard under the name “FT30”.15, 16, 29

Since the rapid development in the 1970s, there have only been small improvements in salt

rejection and fouling resistance in commercial processes, as well as developments for niche

applications such as boron removal and co-contaminants. The primary hurdles of a chlorine

tolerant membrane and a significantly fouling-resistant membrane remain elusive. 15, 16 However,

16

even without resolution of these membrane limitations, the use of RO systems and the number of

applications for which RO is ideal have rapidly increased. Although the primary application is

desalination of seawater, there are also RO plants constructed for the more economically feasible

desalination of brackish water, as well as for treatment of contaminated groundwater and

reclamation of wastewater.29

Reverse osmosis plants have been desalinating brackish water in the Middle East since the

late 1960s, and began desalinating seawater in the late 1970s.15, 30 Due to improvements in RO

technology, it is the fastest growing process for desalination, and in the United States it is the

predominant process. Across Europe, large-scale RO desalination plants have been constructed

rapidly in the last two decades.29 In the Middle East, where the relatively low cost of energy and

high demand for large outputs of drinkable water favored multi-stage flash distillation, MSF still

predominates but RO plant construction is increasing. Reverse osmosis is clearly the desalination

method that will dominate the desalination industry in future years, particularly as more regions

become water stressed and as RO technology becomes more economical and efficient. The use of

reverse osmosis to desalinate seawater for human consumption was one of the first applications of

membrane science, and the development of this application is a growing field of interest today.4,

14 A comparison of RO and other membrane treatment processes is shown in Table 1-1.

A typical reverse osmosis system features feedwater, which can be brackish or seawater,

transferred across a semipermeable membrane at an elevated pressure. As pressure forces the water

permeate across the membrane, the feedwater becomes more concentrated until it is removed from

the system. While operations desalinating seawater generally require a feedwater pressure of 800

to 1200 psi, brackish water desalination requires a much lower and more affordable (energetically

speaking) feedwater pressure of 100-600 psi, and offers obvious advantages where available.5, 14

17

Table 1-1. Comparison of membrane separation processes.5

Process Type Membrane

type

Mechanism Energy

requirement

Application Exclusion

size/MW

Reverse Osmosis

Asymmetric or thin film

composite

Solution-Diffusion

Pressure (5-8MPa for

seawater desalination)

Removal of all dissolved

components

Monovalent ions

Nanofiltration Asymmetric

or thin film composite

Solution-

Diffusion, size

exclusion

Pressure

(0.5-1.5MPa)

Removal of

divalent components

and larger molecules

Divalent ions,

organic carbon, Color, limited

monovalent ions

Ultrafiltration Microporous

Sieves

Filtration Pressure (50-

500 KPa)

Removal of

small microscopic

contaminants

Bacteria and

viruses

Microfiltration Microporous

sieves

Filtration Pressure (50-

500 KPa)

Removal of

microscopic

contaminants

Bacteria and

protozoa,

coagulated organic

particles, precipitated

matter >100 nm

ED/EDR Dense membrane

Ion exchange

Electricity Removal of dissolved

ions

Dissolved ions

1.4.4 Electrodialysis/Electrodialysis Reversal

Electrodialysis/Electrodialysis Reversal (ED/EDR), which has been in use since the early

1960’s, includes the same size range of excluded ions as reverse osmosis, but does not remove

pathogens, suspended solids or any noncharged or nonionic components. 5, 14 As opposed to the

typical transport mechanism in the filtration- or solution-diffusion-based processes where

contaminated water is treated by migration through the body of a membrane, ED consists of

contaminated water passing along or across the surface of parallel ion-exchange membranes. These

membranes selectively remove cations and anions using electrical energy rather than a pressure

gradient, and require both good selectivity for the respective ions as well as a low resistance to ion

18

transfer. 5, 14 The primary disadvantages of ED include a high propensity towards fouling and a

slightly higher operations cost compared to reverse osmosis, but the process also offers the

flexibility of high pH range tolerance (ED is viable from pH 1.0-13.0), high temperature tolerance

at 43 °C, and conversion rates (the percentage of feedwater which is actually converted to product

water) ranging from 50-90%, depending on operating conditions.29

Electrodialysis reversal utilizes the same physical setup as ED, but the polarity of the

electrodes is periodically reversed, along with the wastewater and product water outlets, in order

to reduce the accumulation of scale, biofoulants, and other contaminants. This reduces the need

for cleaning procedures and increases the water recovery potential over a system using ED, but

EDR also requires more skilled labor due to the more complex requirements for maintenance and

operation of the system. 5, 14

Although technically capable of treating seawater, ED/EDR is more typically used for

brackish water, wastewater treatment, and reclamation of components in the food and

pharmaceutical industries. Electrodialysis is only applicable for feedwaters where the total

dissolved solids level is less than 5 g/L as the level of electrical current needed to remove higher

levels of contamination is impractical. Electrodialysis can be used to separate and concentrate

acids, bases, or salts from a dilute aqueous solution. 5, 14

1.4.5 Cost effectiveness and comparative advantages of different desalination

processes

Overall, membrane separations processes are much more energy-efficient than phase-

changing processes due to the significant heats of vaporization and fusion of water. The relative

costs vary according to geography, as some regions such as the Middle East feature both an intense

19

need for drinkable water, with only seawater as a viable feed source, and a very low cost of energy.

The increasingly volatile cost of energy should be factored in when making predictions about the

viability and competitiveness of phase change versus membrane separation-based desalination

plants.11, 12, 50 Furthermore, while thermal distillation methods offer a simple process, reverse

osmosis operates at ambient temperature and relies on non-corroding polymers, which simplifies

the heating and cooling steps as well as reduces the cost of systems monitoring and maintenance.

Additional advantages are the use of elements which can be easily replaced without extensive

training and small industrial footprint. These are not as immediately obvious as the energy use

advantage of reverse osmosis, but nonetheless are significant in the daily operations of a plant.15

Cost concerns include both the capital costs of starting up a plant and the daily operations

cost of producing a given amount of water at a given level of quality, according to levels of

individual contaminants and total dissolved solids.29, 51 Another cost concern which may be more

difficult to quantify precisely is the environmental impact, such as the impact of discharged

treatment chemicals or in the case of RO, brine disposal. A cost comparison of desalination

methods is listed below in Table 1-2.

20

Table 1-2. Comparison of costs for varying desalination systems, in 2003-2006 US Dollars.7, 13, 23, 25, 51, 52

Pretreatment

Requirements

Energy

Required

Cost of

Water

Environmental

Impact

Notes

Reverse

Osmosis

High 2.5-5kWh/

m3

$0.50-

0.70/m3,

seawater

$0.26-0.54/

m3, brackish

water

No temperature

disturbance,

high-TDS

brackish water

disposal

problematic

Values are

for large

volume

plants, over

50,000

m3/day

Multistage

Flash

Low 3-5kWh/ m3 $1.10/m3 Brine

concentrate

features higher

temperatures

Low

product

water TDS

Multieffect

Distillation

Low 1.5-2.5kWh/

m3

$0.80/m3 Similar to MSF Low

product

water TDS

ED/EDR Medium $0.5kWh/ m3 m3 Low Not suitable

for large

volumes of

highly

saline water,

cost is TDS-

based

Reverse osmosis is by far the most versatile and economical desalination technology in

use, and a clear trend towards improved performance and decreased cost in recent years has been

demonstrated, as shown in Figure 1-2. There is a marked economy of scale, as small plants of less

than 5,000 m3/day may cost as much as $1.20-3.90/m3 depending on operating conditions, while

larger plants of over 60,000 m3/day see the cost of water production drop to $0.50-0.70.51

21

Figure 1-2. Power consumption of seawater reverse osmosis has rapidly decreased since its development in the 60s and 70s,

and is now approaching the theoretical minimum.1

1.4.6 Pressure Retarded Osmosis and Forward Osmosis

Both pressure retarded osmosis (PRO) and forward osmosis (FO) are processes in which

energy is produced from the osmotic pressure generated by separating fresh water (or more dilute

contaminated water, sometimes wastewater) from seawater or other concentrated solution by

means of a semipermeable membrane. They are sometimes referred to jointly as osmotically driven

membrane processes. In both processes, the concentrated solution is also referred to as the draw

solution. This is typically seawater although research is ongoing to identify a more competitive

draw solution.

The difference between PRO and FO is illustrated in Figure 1-3. Forward osmosis

harnesses energy directly from the dilution of seawater, while PRO uses a small applied pressure

on the permeate side to reduce the operating pressure to provide more efficient energy conversion.

The reason for the pressure retardation is illustrated in Figure 1-4. PRO is more often used for

22

power generation, and while FO is also a possible power generation process, it is more frequently

used for osmotic dilution or concentration of solutions with minimal energy consumption.53

Pressure retarded osmosis (PRO) is the process by which energy is generated using an

applied force on the brine side of the membrane in order to optimize the efficiency of the work

done by osmotic pressure. PRO, which may also be referred to as engineered osmosis, is unique

among power generation methodologies in that it harnesses energy produced by the entropic

energy of mixing two solutions, and offers a source of renewable energy which features a

negligible impact on the environment. Forward osmosis is a closely related process where no

pressure is applied on the brine side, as illustrated in Figure 1-3.

Figure 1-3. Comparison of forward osmosis, pressure retarded osmosis, and reverse osmosis.

As the energy is generated from a pressure differential, it may seem counterintuitive to

apply a pressure on the permeate side of the process, as is done in PRO. However, when

considering the equations for power density and water flux (the significance of these equations for

reverse osmosis applications will be discussed in Section 1.5), the need for applied pressure

becomes apparent. The following equation describes the water flux:

𝐽𝑤 = 𝐴(∆𝑃 − ∆𝜋)

23

Above, Jw represents water flux, A represents the permeability constant for the given

membrane material, ∆𝑃 represents the applied pressure and ∆𝜋 represents the osmotic pressure

generated between the draw solution and the feed solution. The below equation for determining

power density is given as follows, where W is power density and the remaining terms are as

previously described.54

𝑊 = 𝐽𝑤∆𝑃 = ∆𝑃 ∙ 𝐴(∆𝑃 − ∆𝜋)

Differentiating the power density equation below gives a peak in the power density when

the applied pressure equals half the osmotic power, disregarding A as a material-specific variable.

54

𝑑𝑊

𝑑𝑃= 0, ∆𝑃 =

∆𝜋

2

When ∆𝜋

2 is substituted for ∆𝑃 in the equation for power density, we arrive at a maximum

power density as a function of the permeability constant and the osmotic pressure, thus providing

only two variables to modify for increased power production.54 As the osmotic pressure can only

be modified by the composition and concentration of the draw solute (which is generally seawater

due to its availability and high concentration), this leaves high-flux materials and composites the

primary focus of research in the field of pressure retarded osmosis. It also provides impetus to

develop PRO in areas where highly saline bodies of water offer even higher concentrations than

seawater.55, 56

24

Figure 1-4. Comparison of osmotic and applied pressures for PRO, FO and RO.57

While PRO is not used for production of clean drinking water, the principles and materials

used are very similar to RO and research done on behalf of improving one application is frequently

used in the other. Although FO is also not used directly for the production of clean drinking water,

it has been used in hybrid RO/FO systems, in which FO is used both in the pretreatment of feed

seawater to reduce boron levels and to dilute the product brine generated by the RO step.58

As the primary goal of FO/PRO is sufficient water flux, these processes feature lower

requirements for salt rejection and generated pressure tolerance (0 for FO, 10-15 bar for PRO).

Fouling is also a much smaller concern in FO as the pressure is generated on the ‘pure’ feed water

side of the membrane, with few opportunities for microbes, inorganic components, or organic

material and silt to accumulate on the film surface. The complex maintenance required in RO may

25

be reduced to a simpler occasional physical cleaning or even by careful control of the feed stream

alone (in contrast to a combination of chemical and physical steps required for RO).59

Forward osmosis and pressure retarded osmosis are technologies under development, with

details of efficiency and cost still being evaluated. The primary areas of investigation for FO/PRO

technology are the development of new selective layers,60, 61 surface treatment options,62, 63 and

hydrophilic support layers,64-66 the economics of energy generation,57, 67-72 the efficacy of FO as a

pretreatment or hybrid step in conjunction with RO or nanofiltration,57, 73 and evaluation of various

draw solutes.74-79

Pressure retarded osmosis and forward osmosis processes have an additional challenge of

suffering from concentration polarization on both sides of the membrane, as opposed to the

problem caused only on the feed side in reverse osmosis, which will be discussed in greater detail

in Section 1.6. In PRO and FO, there is the challenge of external concentration polarization, where

water transported from the freshwater feed stream dilutes the draw solution, reducing the osmotic

pressure and therefore causing a sharp decrease in flux. This can be mitigated by thorough mixing

and module design, so the diluted solution moves rapidly across the membrane and leaves the

module before the osmotic pressure is significantly affected. There is also internal concentration

polarization (ICP), where small amounts of solute diffuse across the membrane into the support

layer, reducing osmotic pressure from the freshwater side as well.54, 57, 80, 81 Of the two, the internal

concentration polarization is considered to be the more challenging, as laminar flow of the feed

stream cannot penetrate the porous foam.53, 82-85

While initially the materials used in PRO and FO were identical to those used in RO, some

changes were made to improve water flux, while some decrease in salt rejection was considered

26

tolerable. For example, the nonwoven backing used in thin film composites in RO was removed,

as the lower operating pressures would tolerate lower levels of mechanical support. After removing

the nonwoven backing, the structural resistance to flow was reduced significantly and the flux

increased by a factor of 5-10.53 There are at least three current commercial membranes for

osmotically driven membrane processes available, two of which are thin film composites based

on cellulose triacetate and a third which is undisclosed.53

Modifying the chemical and physical structure of the support foam is also a major thrust

of research, since the direction of water flow is from the foam side towards the dense layer. A

hydrophilic support foam offers potential advantages in minimizing fouling and facilitating

improved water flux.86-88 There is not yet a consensus on what foam structure will be most

advantageous for PRO/FO, be it with or without macrovoids or with an open scaffold or more

closed structure. Macrovoids reduce mechanical strength, but they also reduce internal

concentration polarization and so they may be tolerated within limits. There are also membranes

which feature a very thin dense layer on both sides of the foam structure to minimize irreversible

internal fouling.86-88

There is some controversy in the membrane separations field over whether the benefits and

costs of PRO have been exaggerated.89, 90 However, several analyses have found that at least

2.5W/m2 is a reasonable estimate for power generation,71, 72 and other researchers have suggested

more optimistic projections of 5W/m2 if sufficient improvements in membrane design can be

achieved or if higher concentrations of draw solution become available.54, 80 The latter option may

be viable for plants located at high-salinity bodies of water such as the Dead Sea or the Great Salt

Lake. PRO does offer the advantage of continuously available renewable power, as opposed to the

periodic or entirely unpredictable nature of solar, wind, and other renewable energy sources.71

27

Meeting cost requirements for competitive power generation is more challenging, but may be

possible with improvements in material development.80, 90

Forward osmosis also may be hybridized with reverse osmosis, in which the concentrated

brine produced by RO is then diluted once more using seawater, generating energy to help drive

the RO process. This hybrid system offers the additional advantage of providing a pretreatment

step for the RO feedwater, which may help avoid the necessity of a two-pass system by removing

part of the boron content during the FO step.57 A hybrid system is illustrated in Figure 1-5.83, 91

Figure 1-5. Illustration of hybrid RO-FO system.53

The pressure gradient between two solutions of different concentrations may also be

harnessed in the process of reverse electrodialysis (RED) in order to generate electricity. Reverse

electrodialysis is considered one of the most efficient means for extracting electrical energy from

the thermodynamic energy of mixing solutions.92

While PRO systems predominantly use seawater as a draw solution due to its negligible

cost, there are investigations into four main classes of alternative draw solutes to improve

28

performance: inorganic salts; thermolytic/volatile compounds; organic molecules such as alcohol,

sugar, protein, and organic salts; and polymer-based solutes such as polyethylene glycol,

dendrimers, hydrogels, and magnetic nanoparticles.53 The primary inorganic salt in use is NaCl in

the form of seawater, but overall this class of solutes offers an inexpensive and easily accessible

means of generating a high osmotic pressure in PRO or FO, albeit with a difficult means of

separation or recovery. In the case of seawater this is not a challenge as the dilute water can be

returned to the ocean in an area of brackish water, but for other inorganic salts it remains a

challenge. Thermolytic/volatile solutes are a particularly interesting class, since they may be

evaporated and recovered by using relatively low temperatures generated from waste heat

(particularly power plants). 76, 79, 93, 94

The most studied and promising of these alternative solutes is a solution of NH3/CO2,

which can easily generate much higher osmotic pressures than seawater and can be recovered at

58oC. However, meeting satisfactory recovery levels of NH3 remains a significant technical

challenge. This process is being pursued commercially by Oasys Water. Organic molecules are

not as competitive in terms of generating high water flux, but offer the advantage of easier recovery

than inorganic salts due to a higher molecular weight, as well as being frequently biodegradable

after use or in the event of accidental leaks. Finally, polymer-based solutes offer the advantage of

low-cost and simple solute recovery, most often with UF, as well as easily tailored structures for

the desired osmotic pressure and performance, but are not always easy to obtain or maintain. 53

Pressure retarded osmosis and forward osmosis are both developing technologies and as

such have a number of features which are currently under investigation for improvement.

However, with sufficient improvement PRO/FO may become a commercially competitive entity.

The most critical is improvement in membrane material and structure to get sufficient water flux

29

and structure parameter values,85 followed by controlling internal and external concentration

polarization, and the evaluation of more effective draw solutes. However, currently the use of

seawater, brine concentrate produced from desalination, and the ammonia-carbon dioxide system

utilized by Oasys water are the most viable options.57, 80, 82, 85, 95

1.4.7 Pervaporation

Pervaporation is technically a membrane separation process, but is sometimes also

classified as a thermal process as it involves a phase change. It was first identified as a possible

separations process in 1906, but did not see commercial implementation until the early 1980s with

the application of asymmetric membranes and composites. It is now seen as competitive with, if

not more efficient than, traditional distillation. Pervaporation features a liquid feed stream and a

membrane which allows permeation only to water vapor. The permeate vapor is collected on the

permeate side and then fed to a condenser where it is removed as a liquid, as shown below in

Figure 1-6.96, 97

Although pervaporation is predominantly used to separate ethanol from water in industrial,

agricultural, and beverage production applications, it is also a possibility in the removal of small

amounts of volatile organic compounds (VOCs) from contaminated groundwater or industrial

processes.96, 98 Pervaporation typically relies on a solution-diffusion mechanism through a dense

or asymmetric membrane, but the pressure differential needed is small compared to reverse

osmosis.29, 99

30

Figure 1-6. Schematic of a pervaporation system, in which feed and retentate streams are liquid but permeate stream

consists of vapor.29

1.5 Design of the Reverse Osmosis Process

The reverse osmosis process can be broken down into three primary stages. The first is that

of pretreatment, where the source water is rendered more suitable for treatment in order to

minimize fouling and have a more efficient process. This is followed by the actual membrane

treatment process in RO modules, and then post-treatment, where the produced water is degasified,

and any needed disinfectants or chemicals are added prior to distribution. The process by which

RO produces desalinated water using a solid polymer membrane is termed solution-diffusion

separation. There are a number of terms which will be discussed in greater detail in a later section

used to describe the efficiency of both potable water production, salt rejection, and other

parameters of the RO process.14

1.5.1 Solution-Diffusion Mechanism

In the membrane selection processes, there are two primary mechanisms of separation

depending on the membrane structure. In the pore flow mechanism, a porous membrane excludes

components which exceed the pore size. Porous membranes as a result have very stringent

requirements of maximum pore size and pore size distribution, and the rejection of suspended

31

components such as microbes can be measured on a log scale due to highly successful selection.40

They also produce a higher volume of permeate at a lower pressure than solution-diffusion

membranes, but due to the open structure of the membrane experience a much higher incidence of

fouling.14 In porous membranes, fouling may be external (reversible caking, scaling, or growth on

the membrane surface) or internal (particulates lodged in pores in the thickness of the membrane,

which is largely irreversible).19 In pore flow separation as in solution-diffusion, the driving force

is pressure, but in the case of the former it is specifically the solution properties and the differential

between the entry and exit pores or cylinders and the radii of the solutes or particles of interest

used to model the structure of the membrane.10, 19, 29

In the solution diffusion mechanism, a dense, perfectly pinhole-free membrane separates

one component of a mixture based on the relative solubility and diffusivity of the different

components. Since the permeate component is separated by physical diffusion through a solid,

rather than by passing through a porous sieve, the throughput for a solution diffusion membrane

is much lower than processes relying on pore flow mechanisms. Due to the reliance of reverse

osmosis on solution diffusion mechanisms, this work will focus more heavily on this mechanism

rather than pore flow. 19, 29

Physically speaking, the distinction between the pore-flow sieving model and the solution-

diffusion model consists of the size of the membrane pores. Although the membranes that separate

by means of solution diffusion are dense, they do possess very small pores. While the pore size in

the pore-flow model is an obvious factor, the presence of pores in dense membranes using solution-

diffusion-based selection may seem counterintuitive. However, membranes used to separate by

means of the solution-diffusion mechanism do have very small pores on the scale of 0.5-1.0 nm,

which are created and removed by the thermal motion of polymer chains within the membranes.

32

According to Baker, these pores appear and disappear on a brief timescale, similar to that of the

permeants crossing the membrane. The larger these temporary pores are, the more time they are

present before moving away, providing more of an influence on membrane transport properties

and approaching the behavior of pore-flow membranes. 19, 29

While the equations for the description of a given membrane’s efficacy as an RO membrane

will be discussed in the following section, the basic equation that describes the transport of a given

component across a membrane barrier as a result of a concentration gradient is shown below. In

this equation, Ji represents flux, Di the coefficient of diffusion, and 𝑑𝑐𝑖

𝑑𝑥 the concentration gradient

across a distance x. The negative sign adjusts for the direction of water flow. This is termed Fickian

diffusion and can be used to describe transport of water within the hydrated membrane, excluding

considerations of applied pressure. 19, 29, 100

𝐽𝑖 = −𝐷𝑖

𝑑𝑐𝑖

𝑑𝑥

There is a three-step process to the transport of a solute using the solution-diffusion

process. The first is the initial contact between the solute and the solution/polymer interface, in

which the solute dissolves into the membrane. The second step is the actual transport driven by

the applied pressure across the thickness of the polymer membrane is the primary focus of RO

material research. The final step is the dissociation of the solute from the permeate side of the

membrane. The solution-diffusion model assumes that the first and last steps are much more rapid

than the diffusion-controlled transport, and therefore do not need to be described mathematically.

It is also assumed that the membrane has a uniform gradient across its thickness and that each side

of the membrane is at equilibrium with the liquid or gas at its interface. The solution-diffusion

33

model further assumes that the membrane, when pressurized, evenly distributes pressure within

the system. These assumptions are illustrated in Figure 1-7. 19, 29

Figure 1-7. Illustration of solution-diffusion model, in which p represents applied pressure, i indicates the chemical

potential of dissolved species i, and ici represents the solvent activity of i.29

The critical portion of Figure 1-7 is that of chemical potential, which is influenced by

concentration and pressure (as well as other forces such as temperature, but for reverse osmosis

pressure is the primary factor). As with concentration, a given dissolved species i will move from

higher values to lower values of .29, 37

𝐴 =𝐷𝑤𝐾𝑤

𝐿 𝑐𝑤0𝑣𝑤

𝑙𝑅𝑇

𝐽𝑤 = 𝐴(∆𝑃 − ∆𝜋)

Above, the second equation describes the calculation of A (which is typically empirically

determined, but a reproducible A value can be used to back-calculate the diffusivity coefficient),

in which, Dw is the diffusivity of water, KwL is the sorption coefficient, cw0 is the concentration of

water on the feed or upstream side of the membrane, vw is the molar volume, l is the thickness of

34

the membrane, R is the gas constant and T the temperature in Kelvins. The following equation

models the flux of water, Jw, against a gradient as a result of pressure applied (P) in excess of

osmotic pressure (), with a water permeability constant A.29, 37

The solution diffusion mechanism was not discovered until after the advent of reverse

osmosis membranes. It was heavily disputed during its initial decade of development as initially

pore flow theory was thought to be a better model, as it applied to other membrane separations

applications such as microfiltration and ultrafiltration. Donald R. Paul was instrumental in the

development of the solution-diffusion theory (SDT), which accurately models solute transport

across a dense, pore-free and pinhole-free membrane. Further research has indicated that while

highly effective for describing transport in separations of aqueous solutions, this model does not

apply well to separations of organic liquids. More detailed modeling and calculations have been

performed for those applications.100-102

1.5.2. Terms and Equations in Membrane Transport Theory

The two most important parameters in membrane separation measure the capacity of the

membrane to exclude a given solute, most frequently sodium chloride, and to allow water to diffuse

across the membrane. These parameters can be expressed in several different terms. All equations

given here assume a solution-diffusion model of transport rather than a sieving or filtration

mechanism.19, 37

The calculation of water flux Jw and water transport coefficient was described in the

previous section. Salt rejection may be expressed in terms of salt transport, Js, as determined by

the following equation, were B is the salt permeability coefficient, CBL is the concentration of salt

at the boundary layer (or in the bulk feedwater, if boundary layer data is unavailable or if fluid

35

velocity is sufficient to disregard concentration polarization), and CP is the concentration of salt in

the permeate. The second equation below describes the calculation of the B term, which also may

be observed empirically. In this equation, Ds is the diffusivity of salt in a given material, Ks is the

sorption constant, and l is the thickness of the membrane. It should be noted here that while the

water transport may be increased by applying additional pressure, the salt transport (and

conversely, rejection) is not influenced by pressure.29, 37

𝐽𝑠 = 𝐵(𝐶𝐹 − 𝐶𝑃)

𝐵 = 𝐷𝑠𝐾𝑠

𝑙

In the production of clean permeate water, the recovery term Y is used to describe the level

of conversion from the feed to the permeate, as much of the water remains in the retentate. This is

described by taking the percentage of the ratio between the permeate flux Qp and the feed flux, Qf,

both of which can be measured empirically rather than by calculation. Typical Y values may range

from 40-60%, since excessively concentrating retentate becomes economically unviable in the

separations process when seawater or even brackish water is the feed source. Highly concentrated

retentate also contributes to fouling and scaling as minerals precipitate onto the membrane

surface.15, 37

𝑌 = 𝑄𝑃

𝑄𝑓∗ 100

The structure parameter S is a factor only considered important in membrane systems such

as osmotic power and pressure retarded osmosis, where the critical factor is more the facilitation

of a high water flux rather than a highly selective salt rejection. Since these processes frequently

feature fluid flow from the support foam side of the membrane towards the selective layer, the

36

ability of water to move quickly through the foam and skin is important for good flux data. The

structure parameter measures the structural resistance of the entirety of the membrane towards

water flow, and this parameter also dictates the required performance of the membrane in terms of

both salt rejection and water flux. It is calculated as shown below, in which t represents the

thickness of the membrane, indicates tortuosity, D represents diffusivity and the porosity of the

foam.37

𝑆 =𝜏∗𝑡

𝐷∗𝜀

Given that Cf represents the concentration of a component in the feed, and Cp the

concentration of that same component in the permeate stream, the percent rejection is calculated

below, along with the percent salt flux or passage.

𝑆𝑎𝑙𝑡 𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛 = [𝐶𝑓 − 𝐶𝑝

𝐶𝑓] ∗ 100

𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑆𝑎𝑙𝑡 𝐹𝑙𝑢𝑥 = 100 − 𝑃𝑒𝑟𝑐𝑒𝑛𝑡 𝑅𝑒𝑗𝑒𝑐𝑡𝑖𝑜𝑛 = (𝐶𝑝

𝐶𝑓) ∗ 100

Concentration polarization in a given system is sometimes measured quantitatively in terms

of Beta, a unitless number representing the ratio of a given ion concentration at the membrane

surface to its concentration in the bulk feedwater. A higher Beta number indicates a greater

probability of fouling or scaling on the membrane surface.16

1.5.3 Pre-treatment

Pre-treatment of feedwater is widely regarded as the most critical component of seawater

RO plant design, due to the importance of minimizing fouling in order to optimize membrane

performance and lifetime.31, 38, 103-105 In particular, biofouling requires careful monitoring as the

37

short generation time of microbes results in rapid contamination of membranes. In monitoring the

RO modules, care must also be taken not to add any unneeded chemicals or treatments as they may

damage the membrane or simply cause extra expense both in their application and in their removal

from the feed water. As the concentrations of individual feed water components can vary widely

from one treatment facility to the next as well as through the months of the years, continuous

monitoring of water quality and components is necessary for effective pre-treatment. The primary

concern addressed with pre-treatment procedures is preventing fouling, and the main approaches

consist of identifying and treating the most common types of scaling, particulates, organic material

and microbial organisms which might propagate in the nutrient-rich environment provided by the

first three components.16, 106-109

The proper pre-treatment procedures are usually determined through a combination of

consistent feed water testing and monitoring of the level and types of fouling on the membranes

observed before cleaning, as well as any components that may still be present in the permeate

water. Pre-treatment testing involves identifying both cations and anions which may produce

scaling, as well as variables such as dissolved oxygen which may contribute to microbial growth

in the form of biofouling.4, 16

The pretreatment stage entails the addition of scale inhibitor and acid (typically sulfuric

acid), which increase the solubility of partially insoluble salts (primarily calcium carbonate and

magnesium hydroxide) that might precipitate and foul the membranes. The source water also is

passed through a 5 to 20 m cartridge to remove particulates that may also foul or damage the

membrane. The removal of particulates is more critical for reverse osmosis than for membrane

processes which remove larger contaminants. This is because RO uses a dense membrane rather

than a porous filter, and therefore backwashing to flush any accumulated particulates will not be

38

effective for removing the fouling layer. Fouling is a significant concern in the development and

design for RO systems for this reason, as significant particulate accumulation can quickly damage

the system. Pre-treatment may also entail air flotation or sedimentation, or any number of filtration

processes not including those described above.14, 15, 31, 37

In reverse osmosis as well as multistage flash and some other desalination processes,

chlorine is frequently added at a dosage level of 1-2 ppm in order to deter biofouling. Due to the

potential for rapid deterioration of the RO membrane, the chlorine is removed immediately prior

to the actual treatment process by adding 2-8 ppm of sodium bisulfite or by means of UV

treatment.3, 29

1.5.4 Membrane treatment

In the membrane process, the source water is passed through a series of modules containing

the selective membranes for sodium and chlorine removal. A module design should feature safe

operation under high operating pressures, ease of flushing and maintenance cleaning, minimal

pressure drops, resistance to corrosion, and viability for long-term operation.15 In membrane

testing procedures and limited commercial applications, membranes are operated in dead-end

filtration mode, in which feed water flow is normal to the membrane surface, requiring 100%

conversion of feed water to permeate and rapidly producing a fouling cake layer. This mode is not

viable for high-volume commercial applications, but does offer the advantage of a simple design

and operation for evaluating new membranes. Commercial membrane separation almost always

features cross-flow filtration, in which the feed water flows over the membrane surface rather than

being pressurized directly against it. This design reduces opportunities for fouling as the pressure

parallel to the membrane surface inhibits deposition and adhesion of particulates and microbes.15,

16, 37

39

While original membranes were developed in the flat sheet or so-called plate and frame

configuration, this was an expensive and inefficient method of membrane separation. Modern

modules typically use spiral-wound flat sheet membranes or hollow fibers. See Table 1-3 for a

comparison of flat sheet, spiral-wound and hollow fiber element properties, and Figures 1-8 and

1-9 for an illustration of each type of module.

Table 1-3. Comparison of module types used in reverse osmosis.15, 16

Module Type

Advantages Disadvantages Fouling Potential

Cost Notes

Flat sheet (plate and

frame)

Limited fouling

Low surface area to volume

ratio

Moderate High No longer used commercially

Spiral Wound

Easy to handle and clean,

fouling resistant,

widely

available

Some concentration

polarization, low recovery

value Y

Moderate to high

Moderate Mostly crosslinked

polyamide composite

Hollow

Fiber

High surface

area to volume

ratio

Fouling

problematic,

limited variety

in supply

Extremely

high

Moderate Mostly

cellulose

Acetate or

asymmetric

aramids

Spiral-wound elements are produced from flat sheets of membrane, folded over a permeate

stream space to produce a long envelope which is glued along both ends of its length such that it

is only open on the feedwater end, and wound around a pipe to collect the product water. Feed

water is introduced to the outside of this envelope and flows along its length, thus reducing fouling

by maintaining pressure parallel rather than normal to the surface, and the concentrated brine is

collected at the end of the envelope from the feedwater spacer.5 Spiral wound elements have the

advantage of being resistant to fouling due to the open feed channels, and are easy to clean when

40

fouling does impair performance. They are widely available, easy to handle and may be replaced

without significant interruption to the overall system.15

Figure 1-8. Illustration of a spiral-wound RO module.

Hollow fibers, also referred to as hollow fine-fiber elements (HFF) in order to distinguish

them from the hollow fibers with a larger diameter used in microfiltration and ultrafiltration, are a

module type where a bundle of hollow fibers are contained within the module element. Pressurized

feed water is typically introduced via a pipe in the center of the element, and water passes radially

from the outside of the hollow fibers toward the inside, where it flows outside of the module to be

collected. In Figure 1-6, a schematic of the hollow fiber unit illustrates that the fibers feature both

ends of the fibers fixed in the permeate-collecting side of the module, while at the opposite side,

the folded bend of each fiber is held in place. Hollow fibers offer a much higher active surface

area per module in comparison to spiral-wound elements, but are much more susceptible to fouling

and feature a lower flux than spiral wound elements. During the treatment process, the membranes

may require cleaning, either through chemical cleaning (to combat fouling) or backflushing the

system (to remove scale and biofilms with or without the additional application of chemicals).5

41

Figure 1-9. Schematic of a hollow fiber RO element.

1.5.5 Post-treatment

The post-treatment phase consists of chemical remediation and disinfection of the product

water. This entails pH remediation, most commonly by elevating an overly acidic permeate with

lime or sodium bicarbonate, as well as decarbonation or degasification of the treated water to

remove carbon dioxide produced during the acidification process. This is necessary as membranes

do not generally reject carbon dioxide and when allowed to remain in the product water it can

contribute to excess acidity and/or deplete the anionic resin used in ion exchange treatment at the

end of the post-treatment process.15, 29 This is followed by the addition of disinfectants and

stabilizing chemicals in order to ensure the water remains drinkable by the time it arrives at its

intended location.15 The most common additive is chlorine or a chlorine-containing compound, in

order to deter microbial contamination. Another step frequently used is ion exchange, sometimes

referred to as ion exchange polishing, for the removal of any residual ionic groups, boron or

organic molecules that may still exist in the system.15, 110-113

1.6 Challenges in reverse osmosis

42

While the objective of reverse osmosis is relatively simple, the modeling and prediction of

RO performance in a given membrane can be greatly hindered by the problem of fouling. While

fouling is largely controlled by means of pre-treatment of the feed water, it remains an issue with

even the most thorough pre-treatment procedure. Concentration polarization is an inherent

problem in reverse osmosis and any membrane separation procedure, and can be a significant

contributor to several categories of foulants. In addition, there are concerns of being able to remove

dissolved contaminants other than monovalent and divalent salts, specifically boron, and these can

offer distinct challenges to RO systems which are largely developed with the intent to exclude

sodium, chloride, magnesium, and other major components of seawater. Another area of research

is solutions to the problem of brine disposal, which is the primary issue prohibiting RO from being

considered a more environmentally friendly option.

1.6.1 Fouling

Fouling is the decline in water flux due to the accumulation of solids on the membrane,

particularly in the pores of a porous membrane or on a rough surface produced by a crosslinking

process.21 Internal fouling of the pores is effectively not a problem in RO membranes, and so for

the purpose of this application only surface fouling will be discussed. These solids may consist of

biological cultures of algae or other microbes, requiring treatment with chemicals which will not

only kill but detach the cell walls or adhesive gels from the membrane surface without damaging

it physically, or they may be scales of inorganic salts which have precipitated on the surface. The

two primary complications caused by fouling are: an excessive operating pressure to maintain

sufficient flux in spite of accumulated fouling layers, and an increased pressure drop between feed

and permeate streams due to the barrier layer formed by fouling. These are both due to the

decreased water flux since the fouling layer physically impedes the contact of feed water with the

43

membrane. Fouling can also decrease membrane performance by physically damaging the thin

selective layer via abrasion or other mechanical degradation. Fouling is the primary problem in

operating a sea water reverse osmosis (SWRO) facility, as it reduces module lifetimes and reduces

flux while increasing maintenance costs.

While most fouling is partially reversible, there is generally a small amount of performance

lost to irreversible fouling even with regular backflushing and chemical cleaning. Likewise,

membrane modules will need to be replaced either due to irreversible fouling or membrane failure

due to accumulated fouling. Fouling may also be minimized by the selection of membrane

materials which are smoother or offer lower surface adhesion to fouling components. The

development of surface treatments or novel materials which provide a greater resistance to fouling

is a significant area of research today.106, 114-121

Concentration polarization is a frequent precursor to fouling cake formation, and occurs

when both water flux and salt rejection decrease due to the higher concentration of rejected solutes

near the membrane-permeate interface. Fouling can also reduce the rejection of trace contaminants

other than the target salts.122 There are four primary categories of fouling: suspended particulates,

inorganic scale which precipitates from dissolved salts, organic material, and microbes which form

colonies that grow rapidly in the presence of inorganic and organic components collected on the

membrane surface.37 The first three are largely preventable by proper pretreatment procedures and

careful monitoring of feedwater composition, while biofouling requires both pretreatment and

cleaning procedures in order to be properly controlled.

1.6.1.1 Suspended Particulates and Colloids

44

Suspended particulates such as silt and other largely inorganic solids present some physical

hazard to RO membranes, and can effectively be removed by a pretreatment step of MF/UF, in

which case they do not require remediation or cleaning during the RO treatment process. These

particulates cover a wide range of sizes, from 10 nm to 10 m, and may sometimes require

filtration at several different scales to sufficiently reduce accumulation.29

The level of pretreatment to be included for a given system is determined by calculating

the silt density index (SDI) for the feed water. The SDI test consists of filtering feedwater through

a 0.45 mm microfiltration membrane at 30 psi, where an initial sample and a final sample are

separated by a test time. The SDI is determined by the following equation, in which T i represents

the time to collect the initial sample of 500 mL, Tf represents the time to collect the second sample

of the same size, and Tt represents the total time required for the experiment.19

𝑆𝐷𝐼 =100(1 −

𝑇𝑖𝑇𝑓

⁄ )

𝑇𝑡

A very low SDI of less than 1 indicates that no pretreatment step is necessary and that the

system may run for several years without cleaning, and SDI values of up to 3 indicate that more

frequent cleaning is necessary but still cost effective. In the range of 3-5 indicates that either

frequent cleaning or moderate pretreatment is necessary, and above 5 there must be sufficient

pretreatment steps to reduce to the SDI of the feedwater before it encounters the RO module.19

1.6.1.2 Inorganic Scale

Poorly soluble inorganic salts are prone to precipitating and forming scale during the

reverse osmosis process. This is particularly pronounced in systems with a high level of recovery

and therefore a high degree of feedwater concentration, as well as systems with significant

45

concentration polarization problems. Scale is normally well-controlled in the RO system if there

is sufficient monitoring of feedwater quality and proper pre-treatment procedures and chemicals.15,

123, 124

Calcium carbonate and to a lesser degree, calcium sulfate, phosphate, and silica species are

the primary culprits in fouling from the precipitation of inorganic scale. This is less of a problem

in seawater RO than it is in brackish water RO, due to the lower recovery available in SWRO

which prevents precipitation during concentration polarization.4

Scale accumulation occurs in three stages, the first two of which are reversible. First, the

organization of cations and anions into loose associations, followed by the nucleation and

alignment of ionic groups as ions move closer to each other; and finally the precipitation and

growth of salt crystals around the nuclei.4 Scale is prevented by the addition of antiscalant

compounds during pre-treatment. The use of antiscalant requires some caution, however, as they

are both expensive and when used excessively can contribute to fouling themselves, particularly

in the presence of other non-target contaminants. When pretreatment has not been sufficient to

prevent scale accumulation on membranes, a combination of acidic and basic cleaning chemicals

may be used to remove the accumulation.29, 125-130

1.6.1.3 Organic Material

Organic material is primarily an issue in RO treatment of wastewater, due to the high levels

of organic particulate contaminants. Organic foulants, which consists mostly of humic acids,

decomposed organisms and waste products, are sometimes referred to categorically as natural

organic matter (NOM). Pretreatment with flocculation and microfiltration or ultrafiltration is

usually sufficient to prevent significant performance loss, and in the event of caked accumulation

46

it can be reversed by backwashing. NOM contributes to fouling both directly by forming cakes of

material on the membrane surface and by providing a source of nutrients for the more significant

problem of biofouling by microorganisms. Hydrophobic membrane surfaces are more favorable

for NOM adsorption, and so investigation of more hydrophilic surfaces as a means of avoiding

both NOM fouling and biofouling is an area of investigation for combating the problem of

fouling.131-134

1.6.1.4 Biofouling

Biofouling is the most challenging type of fouling to prevent and remove, as given an

environment with sufficient nutrients; a single microorganism can rapidly form a large colony.

Biofouling microorganisms tend to be oligotrophs, or organisms which are specifically adapted to

thriving in an environment where few nutrients are available. While sufficient monitoring and

pretreatment of the first three types of fouling will help limit biofouling, the cleaning steps required

during occasional maintenance service are largely focused on maximum removal of microbial

colonies on the membrane surface. Biofouling is controlled or minimized by water pretreatment,

through microfiltration and ultrafiltration, by the addition of antimicrobial chemicals to the

feedstream, and by the development of fouling-resistant membranes.135 It is much more difficult

to remove a microbe or young colony which has already adhered to the membrane surface than it

is to deter adhesion prior to any form of attachment, particularly as even a small colony forms a

gel-like area to which other foulants may easily adhere and serve as nutrients for further growth.

A colony may become irreversibly adhered after several days.15, 16, 106-108

Biofouling consists of microbes which have accumulated on and attached to the surface,

and a gel-like biopolymer slime produced by the bacteria. This slime may be referred to as the

glycocalyx, slime layer, capsule, or extracellular polymeric substance (EPS). It has been argued

47

that the bulk of flux decline due to biofouling is caused by the EPS rather than the accumulated

bacteria.5, 14 The primary method of preventing biofouling is adding chlorine as a biocide, but in

the two primary commercial membranes, cellulose acetate and crosslinked polyamide, chlorine

causes chemical degradation of the polymer and loss of performance. Therefore, chlorine must be

removed from the feed water prior to exposure to the membrane. Chlorine removes the EPS before

actually damaging the microbial colonies, and microbes secrete additional EPS in response to

ambient chlorine.15

In the presence of biocide, microbial colonies may be dead or greatly reduced in living

numbers, yet still present a physical barrier to water flux and an adhesive mass where other

nonliving foulants may become lodged. Ozone is another biocide which is more versatile than

chlorine and almost as effective, yet still may pose an indirect risk to chlorine-sensitive materials

as it may oxidize bromide present in the seawater feed, which may undergo a similar but slower

reaction with polyamides.15

Surface treatments predominate over the evaluation of new materials in the literature, as

biofouling is more easily avoided by the inclusion of more hydrophilic components in or on the

membrane surface. Finding a new material that is competitive with commercial membranes is

more difficult. Developing biofouling resistance has been studied primarily by surface treatments

which modify the very upper portion of the selective layer, or by polymerizing a thin layer of low

hydraulic-resistance hydrophilic material such as poly(ethylene glycol) on top of the selective

layer.62, 136-141

Another objective is the formation of a smoother membrane surface. A rougher topology

is created during the crosslinking of the aromatic polyamides most frequently used in commercial

48

modules, which can provide a larger surface area for foulants to lodge. On rough surfaces, the

initial microbes may adhere themselves and be relatively sheltered from the shear force of the

pressurized feedstream.142 In forward osmosis and pressure retarded osmosis membranes, fouling

is a less significant concern but new materials in the support layer or across the asymmetric layer

are still of interest.60, 62, 143

1.6.2 Concentration Polarization

Concentration polarization is an obstacle in reverse osmosis caused by the disparity

between the rate of convective flow in the feedwater, and the much lower rate of diffusive flow of

water molecules across the thickness of the membrane. As feed water moves along the length of a

module bordered by two active membrane surfaces, there is a boundary layer at either surface

where the flow is slower and less mixed. This is shown for one surface in Figure 1-10, in which

Cs, bulk represents the concentration of the solute in the bulk phase of the feed, Cs,0 represents the

elevated concentration near the membrane surface, and Cs,l represents the concentration of the

permeate. Due to the high level of ion rejection and low level of fluid movement, the boundary

layer next to the membrane surface features a much higher concentration of the dissolved solute(s)

than that of the bulk feedwater. Concentration polarization impedes performance of the reverse

osmosis membrane by not only providing physical obstruction to feedwater flow, but also by

increasing the osmotic pressure . This reduces the driving force for producing water flux, and

sometimes increases the flux of ions across the membrane.16, 29

49

Figure 1-10. Illustration of concentration polarization.10

Concentration polarization can result in either a decreased water flux due to a disrupted

water gradient near the membrane surface, decreased salt rejection due to an increased ion gradient,

or both. Concentration polarization can with time also produce gels or cakes from precipitated or

agglomerated species of foulants as discussed previously.144 Aside from careful pre-treatment of

the feed water and regular testing and inspection of RO modules, concentration polarization can

be minimized by good mixing of the feed solution and a high velocity parallel to the surface to

disrupt accumulated solute.3

1.6.3 Removal of heavy metals

In some regions, water which is technically fresh (less than 1000 ppm TDS) may still not

be drinkable due to heavy lead, arsenate, or other heavy metal content.145-149 In this case, materials

with a high selectivity towards heavy metals are of interest for producing potable water or for

remediation of industrial wastewater prior to discharge.150, 151 While research in this area has not

50

received the same level of attention as the desalination of freshwater and brackish water, as more

areas become water stressed, there may be increased reliance on underground sources that may be

contaminated, and as more industrial pollution occurs, it is likely to be an important area of

development for future research. Arsenic III and V in particular are contaminants of concern that

have been investigated with conventional RO membranes, with the more promising rejection

values at higher pH levels.147

1.6.4 Removal of boron

Boron, which generally is found in seawater in the form of boric acid, presents a unique

challenge for membrane separations research. While boron at very low levels is beneficial and

possibly essential for humans, it frequently occurs in higher levels in both seawater and brackish

water that exceed WHO standards and is considered detrimental to human health. Boron is difficult

to remove via membrane separations due to the resistance of boron compounds to forming ions

unless they are in an alkaline environment. While raising the pH does allow for practical boron

removal, additional problems are created by causing increased fouling via precipitated magnesium

and calcium salts, and the pH must be lowered again to be fit for human consumption.144, 152

Boron removal, particularly in older RO facilities, generally entail repeated passes through

RO modules which decreases productivity and increases operations cost. After many years of poor

results, there are some initial reports of single-pass boron rejection exceeding 90%. Research is

ongoing to find a practical solution to removing boron, although some efforts suggest that the most

cost-effective solution at present may entail passing product water through an ion exchange unit

tailored specifically for boron removal.49, 110, 153

1.6.5 Sustainable power use

51

Historically RO plants have been driven using power derived from the surrounding energy

grid. However, due to increasing concerns over the cost and availability of conventional energy

sources there are growing interests in research to provide clean water using alternative energies

with low or no environmental impact and a limitless, if interrupted energy supply. This frequently

involves design plans which integrate solar, hydroelectric, biomass, geothermal and/or wind

supplies with or in place of conventional fossil fuel energy sources. In order to maximize the

production capacity of these sources, there is also research into further improving the efficiency

of elements in the reverse osmosis process (i.e. pumping and energy recovery from heated output

water), although the membrane treatment process itself is already closely approaching its

theoretical minimum efficiency.1, 57 Investigative studies into the possibility of combined

alternative energy and desalination technologies emphasize the long-term nature of these

investments, but the inevitable increase in the cost of fossil fuel-based energy requires a thorough

investigation into future alternatives.31, 50, 154

Solar energy has been a strong historical energy source for the desalination of water, either

by harnessing it directly for purification via distillation or for the conversion from thermal to

electrical energy in order to drive a membrane process. In regions which receive high levels of

sunlight, passive solar energy is already used in part of the waste disposal procedure to evaporate

residual water from the produced brine in order to produce sea salt, as well as integrating solar

energy panels to supplement plant energy costs.69, 155-157

Wind energy is another favorable renewable energy source for supporting RO plants, as

there is a great potential for the coincident development of offshore wind farms, which could power

in part or entirety a seawater desalination plant. Some initial studies have indicated that the most

52

feasible option under current circumstances is using a wind farm as back power support for a plant

which draws most of its energy from conventional sources.31

Wave energy is also a convenient energy source for an RO plant, as both would be

conveniently located for immediate energy use. Wave energy also offers the advantage of being

consistent and predictable in timing, due to its dependency on the tides. Hydrostatic pressure is not

an energy source, but it is possible to reduce the energy costs of submarine RO units by using the

pressure of surrounding seawater. This saves energy both by eliminating the energetic demand of

pumping in feedwater and by using the hydrostatic pressure to drive the production of freshwater

at atmospheric pressure.31, 158

1.6.6 Brine Disposal

One of the ongoing challenging in the establishment of SWRO as a viable water production

technology is addressing the environmental impact of brine disposal, which has been a challenge

since the first desalination plants were constructed. Brine is not only an environmental hazard due

to its high salinity, but also due to the introduction of pretreatment chemicals, cleaning chemicals,

and other contaminants which may be incurred during the desalination process. While the

environmental impact is considered to be detrimental, the precise impact on seawater quality and

local ecology is not well studied and requires further investigation. A number of ideas for brine

disposal have been proposed, including dilution with treated wastewater, transport and dilution far

offshore, injection into empty mines, and potential commercial applications from evaporated brine.

The primary solutions currently used for desalination plants include dilution with industrial

wastewater, conversion to sea salt in on-land evaporation ponds, and offshore dilution into ocean

waters.159-161

53

One promising candidate for resolving the brine disposal issue is the construction of a

hybrid FO/SWRO plant as discussed previously in Section 1.4.7. This not only provides a means

of diluting and recycling brine, but also provides a preliminary pass for removing boron and other

solutes which are challenging to separate at high levels. However, it is a limited solution as

practically speaking, since only part of the brine can be added to the feedwater, and so some must

still be disposed of safely.57

1.7 Membranes and composites used in reverse osmosis and osmotic power

1.7.1 Materials used in reverse osmosis

The high pressures, threat of biological and chemical degradation, and high performance

standards for reverse osmosis membranes provide a very demanding environment for the materials

used in these membranes. An ideal membrane features high water flux, high salt rejection,

resistance to chlorine and other oxidizing chemicals, resistance to biological attack and fouling,

resistance to fouling by colloids and suspended materials, resistance to compaction during use, a

low cost of synthesis and fabrication, ease of fabrication into thin films, asymmetric membranes

or hollow fibers, good mechanical properties, chemical and hydrolytic stability, and tolerance of

high temperatures. While no membrane yet features all of these qualities, several exist with specific

advantages for different applications. While no novel membranes have entered the commercial

market in several decades, investigations are ongoing in the development of new materials that

may overcome the advantages of the current commercial membranes, primarily in the areas of

improved transport behavior and increased fouling resistance.5, 14, 15

Most modern reverse osmosis RO systems utilize either an asymmetric membrane or a thin

film composite, or TFC. The thin film composite consists of a polysulfone foam supported by a

54

tough nonwoven fabric, generally polyester, for mechanical strength. On the foam support, a a thin

film of cellulose acetate (CA) or interfacially polymerized crosslinked aromatic polyamide (CPA)

serves as the selective layer. The TFC was first developed in the early 1980s by FilmTec

Corporation, which is presently a part of Dow Chemical Company, and Fluid Systems, which is

now a part of Koch Membrane Systems.14

1.7.1.1 Cellulose Acetate

Cellulose acetate films were initially invented in 1960 by Loeb and Sourirajan.162 CA films

are typically asymmetric membranes produced from cellulose diacetate, cellulose triacetate, or a

blend of the two membranes. They have a dense layer thickness of 0.2-0.3 m and an overall

membrane thickness of about 100 m, and are supported by a highly porous substrate. Their

primary advantage are a low materials cost and easy availability, and a greater tolerance to minor

chlorine exposure than their crosslinked polyamide competitors. Their comparatively smooth

surface offers some resistance to the adhesion of fouling. However, they are vulnerable to

compaction, which results in the flux that is lower than polyamide membranes, in part due to the

thicker selective layer. More critically, chemical and biological degradation can occur over their

performance lifetime, which may be shortened at high or low pH levels or high temperatures.16

Cellulose acetate asymmetric membranes can operate continuously at pH levels ranging

from 4.0 to 6.5 and up to 30oC, which is slightly cooler than what crosslinked polyamides can

tolerate. This is important due to the increase in membrane flux and overall capacity with elevated

temperature. They were primarily used during the early development of reverse osmosis systems,

but have become less common with the invention of crosslinked polyamide. Currently CA hollow

fibers occupy less than 10% of the RO/NF market, largely by virtue of their chlorine tolerance and

lower cost. They are also more frequently found in brackish water desalination systems.14, 20, 49

55

1.7.1.2 Crosslinked polyamide

Crosslinked polyamide (CPA) films are the predominant commercial RO material in use

at present, occupying over 90% of the RO/NF market.49 Primarily used in TFC membranes in the

late 1980s, CPA shows improved salt rejection over the CA films, with a steady improvement in

flux over the following decade of development.20 Although they have a greater stability over a

wider pH range and are resistant to biological degradation, they are very sensitive to degradation

by chlorine, and are ideally not used at temperatures higher than 45 oC.5, 14, 163, 164 The increased

surface roughness due to the crosslinking reaction provides greater surface area for adsorption of

water, but also create dead zones where foulants may easily accumulate without being disturbed

by the feedwater flow.16 Due to a much thinner selective layer (40-100 nm) than CA membranes,

CPA-based films require a lower operating pressure in order to achieve a satisfactory water flux.

Research into polyamide materials which may be more tolerant of small amounts of chlorine

exposure has yielded some results, but nothing which has been adopted to replace the CPA

standard.165

1.7.1.3 Alternative membranes and materials for SWRO

Although the CA and CPA films comprise most of the commercial market, there are

investigations underway to develop alternative materials. One approach is to interfacially

polymerize a copolymer with polyamide, such as polyurea or polyurethane in a thin film composite

(TFC) with the intent of obtaining a higher salt rejection and higher fouling resistance.166, 167 Other

approaches have included the incorporation of lyotropic liquid crystals (LLCs), carbon nanotubes,

or nanoparticles in a polymer membrane in order to offer improved fouling, flux and rejection

characteristics, but these membranes have not yet shown commercial viability.16, 4930

56

Alternative polymers investigated for RO applications include aromatic systems such as

sulfonated polysulfones and sulfonated poly(ether ether ketones) and related materials, which offer

advantages in chemical resistance and mechanical properties.168, 169 Other materials such as

polyureas and sulfonated polyethers are under commercial development by Nitto Denko.15 Surface

modification techniques for both conventional and novel membrane materials are also under

investigation in order to augment chlorine resistance, water permeability, and fouling resistance.65,

142, 170-172

1.7.2 Materials used in pressure retarded osmosis/forward osmosis

Due to the lower pressures used in osmotic power, a greater tolerance is available for the

construction of osmotic power membranes in terms of accepting a lower salt rejection so long as

the membrane provides a higher flux. The membrane can also be less resistant to compaction and

less tolerant of fouling, given some offset advantage in another property, as the direction of flow

from low to high concentration provides less opportunity for fouling. As forward osmosis is a

developing technology, the same CA and CPA membrane materials which are used in reverse

osmosis are used in FO, although other materials are being explored. Unlike RO membranes,

potential FO candidates do not need to show the same resistance to oxidizing agents and will be

exposed to lower pressures, allowing for greater flexibility in both membrane composition and

thickness.57, 70, 173 This has resulted in the exploration of polybenzimidazoles and polyamide-

imides with specific PRO-FO applications.64, 174, 175

An additional topic of research which is more of interest in osmotic power applications due

to the hazard of internal concentration polarization is the development of a hydrophilic support

with a thinner layer, resulting in a lower S or Structure parameter. As a lower S parameter results

57

in reduced internal concentration polarization, sulfonated materials are of interest in forming these

support layers, either for a thin film composite or asymmetric composite.64, 87, 176-178

1.7.3 Asymmetric membranes and Thin Film Composites

Due to ongoing competition to increase production of drinkable water, even dense

membranes with excellent water permeability are generally insufficient for commercial use. They

are studied in the early stages of research to determine a given material’s transport and rejection

properties, but at the commercial level these materials may be fabricated into thin film composites

(TFCs) or asymmetric membranes (ASMs). This hybrid structure allows for an increased flux due

to a thinner selective layer while maintaining ideal salt rejection values. The structures of dense

membranes, thin film composites, asymmetrical membranes, as well as the microporous isotropic

membrane used in microfiltration, ultrafiltration, and in support foams are illustrated in Figure 1-

11. While currently TFCs are the primary option for SWRO modules, ASMs are still used in some

niche applications. Both types of membranes are used in other related membranes separations

processes, such as pervaporation, forward osmosis, and gas separations.29

58

Figure 1-11. Comparison of membrane structures used in different separations processes.19

Thin film composites (TFCs) are comprised of three layers, generally made from three

different polymers: a highly porous hydrophobic nonwoven paper (120-150 m thick) for

mechanical support (generally polyester), a porous, isotropic or anisotropic foam (40-60 m thick)

for direct mechanical support of the selective layer (polysulfone), and the selective layer itself

(cellulose acetate or a crosslinked polyamide), which is typically 200 nm thick.49 The selective

layer is most often applied by interfacial polymerization on top of the support foam, but also by

solution casting or plasma polymerization.29 This composite system offers several advantages,

most notably that it is able to withstand significant pressures during RO operations. It also allows

59

for selection of different materials to optimize performance in the support foam as well as the

selective layer. Unfortunately, the complex multilayered structure and specifically its hydrophobic

components result in a lowered permeability and diminished structure factor.3

A simpler version of the TFC, and the first of the two to be developed, is the asymmetric

membrane which consists of two layers: the selective layer, generally less than one micrometer in

thickness, and the substantially thicker supportive foam layer of 50-100 microns, which provides

mechanical strength. The foam layer may occasionally feature an isotropic foam structure, where

pore size is largely uniform from the top to the bottom of the foam layer, but more frequently it

features a gradient where pores become larger and more open (sometimes referred to as a scaffold

structure) moving away from the skin layer. These two components are made from the same

copolymer source and are fabricated simultaneously using phase inversion.14, 15

The primary limitations of asymmetrical membranes are due to their source material. For

example, cellulose acetate and related materials commonly used in commercial asymmetric

membranes today face short lifetimes due to hydrolysis, which is more rapid at elevated

temperatures and pH levels outside of 4.5-6.5. However, all asymmetric membranes tend to suffer

from a loss of flux due to compaction in the support foam layer, as well as the concern of screening

out pinholes and other flaws during fabrication. However, due to the simpler production and

uniformly hydrophilic composition of the membrane material, an asymmetric membrane offers the

advantage of a competitive structure factor and a potentially increased flux over a thin film

composite of comparable dimensions.14, 20, 37

1.7.4 Phase inversion

60

Asymmetric membranes are produced by the phase inversion of a polymer solution. Phase

inversion can be performed using a variety of techniques, including thermal inversion and vapor

phase inversion, but the primary method used in RO membranes and to be discussed here is that

of nonsolvent immersion, also called immersion precipitation, which entails the submersion of a

thin film of polymer solution into a bath of nonsolvent liquid.19, 162, 179

The immersion causes very rapid precipitation of the polymer from the top (skin) layer

downwards, and the single-phase polymer solution becomes a two-phase solid, the polymer-rich

dense skin layer and the polymer-poor support foam. Because the formation of the ASM depends

on the kinetics of solvent and nonsolvent transport as well as the thermodynamics of the solution

itself, being able to produce and reproduce ASMs of a desired structure is difficult. While there

are models and trends for the overall process, working with a given material and

solvent/nonsolvent system requires extensive empirical work rather than mathematical predictions.

The following discussion will address the phase separation process both from the stages of

precipitation, and from the impact of the precipitation kinetics on the final membrane structure.19,

29

Phase inversion via precipitation immersion, also referred to as the Loeb-Sourirajan

process, first takes place by means of spreading out a thin layer of polymer solution on a casting

surface. This solution is allowed to stand undisturbed for a period of time ranging from several

seconds to several minutes, during which solvent evaporation from the top of the solution creates

a higher polymer concentration region in what will shortly become the polymer-rich phase of the

ASM. After this wait time, the solution is immersed in nonsolvent (typically water), and the

precipitation commences. Skin thickness and coherence are strongly influenced by polymer

61

concentration, evaporation time, solution and air temperatures as well as the volatility of the

selected solvent.16, 180

While the process of developing a phase inversion protocol is largely empirical and

requires a lengthy period of trial and error, there are several guidelines provided by Baker in

Membrane Technology and Applications that may be of use. The first guideline is that the polymer

being used is amorphous and tough, and has a high molecular weight, preferably at least 40,000

g/mol. Polymers which are semi-crystalline or rigid glasses will be too brittle after casting and

susceptible to failure during processing. The solvent used is ideally an aprotic solvent with a high

solubility in water (the most common nonsolvent used in phase inversion), such as dimethyl

formamide, dimethyl acetamide, and N-methyl pyrrolidone. These solvents will dissolve a wide

range of polymers and when immersed, tend to provide an anisotropic membrane with a good

porosity and pore gradient. The ideal nonsolvent is water, not only for affordability and

convenience but because membranes of comparable material cast in organic nonsolvents such as

acetone or isopropanol have a tendency to show lower fluxes and greater membrane density. In

the situation of a material where water is not a good nonsolvent, some adjustments can be made.

The temperature of the nonsolvent also strongly affects ASM structure and performance, which in

the case of water a reduced temperature produces lower flux and improved salt rejection.19, 181

According to Baker and Wang et al., during the phase inversion process there are four

distinct regions of polymer phases, as indicated by their region number in Figure 12: Region I is

the single-phase polymer solution prior to any precipitation. The time spent going from the single-

solution phase to the liquid-liquid two-phase solution in Region II is called the delay time, and it

is the precursor to the precipitation of the solid polymer. Region II consists of two liquid solutions,

one which is polymer-rich and will form the polymer skin, and the other which features a more

62

dilute polymer concentration. Region IV consists of a glassy polymer with small components of

solvent, and is not typically part of a successful phase-inversion process.19, 29

The boundary between the two regions is termed the binodal boundary, and C is the C-

point, or the “critical coagulation point”, which is the ceiling concentration of solvent in a

nonsolvent bath that will still precipitate a solid polymer. For the fabrication of asymmetrical

membranes with good mechanical strength, the nonsolvent concentration must exceed C in the

nonsolvent bath, as it is the boundary between having two liquid solution phases and merely having

a highly swollen polymer gel. That point sets one end of the gelation boundary between the C-

point and point B, which refers to the so-called Berghmans’ point. The Berghmans’ point is located

at the intersection of the swelling boundary and the gelation boundary, and indicates the point of

vitrification of the polymer-rich phase.19, 29

Figure 1-12. Three-phase system describing precipitation inversion for producing asymmetric membranes.19, 29

63

Within Region II, there are lines indicating binodal and spinodal curves, and it should be

noted that compositions between the spinodal line and the gelation boundary will be unstable, and

that the final porous membrane formed from those compositions will lack mechanical integrity.

Those compositions which fall between the binodal and spinodal curves are considered metastable

and therefore able to withstand small stresses while in transition to the final, more stable structure.

This may be visualized in practice by precipitating a low-concentration solution and a high-

concentration solution. If the former is sufficiently dilute, it will precipitate in the unstable region

below the critical point and upon drying, producing a brittle polymer that fragments upon handling.

Region III indicates the onset of physical precipitation of the polymer-rich phase into a swollen

solid polymer, and the polymer-poor phase becomes a mixed-liquid system of both solvent and

nonsolvent. Region IV represents the final and fixed phase of entirely solid polymer, with a

vitrified skin layer and swollen foam support.19, 29

While the boundaries illustrated in Figure 1-12 are useful for a qualitative understanding

of phase relationships and the immersion precipitation process, they can be quantified and

described mathematically. An important part of initial evaluations of polymer/solvent/nonsolvent

systems is the determination of cloud point correlations. Cloud points are determined by adding

small amounts of nonsolvent to a polymer solution and recording the point at which the solution

remains cloudy after stirring, indicating the approach of precipitation as the polymer is no longer

entirely soluble. These cloud points can be used to qualitatively approximate the phase boundaries

for a given polymer-solvent-nonsolvent mixture and phase inversion process.19, 29

When considering the impact of kinetics on phase separation as depicted in Figure 1-13,

the interplay both between the rates of diffusion of the nonsolvent and the solvent are important

for the final membrane structure. The precipitation of the polymer-poor phase below the initially

64

precipitated skin, or polymer-rich phase, will inevitably be a slower process as both solvent and

nonsolvent must be transported by diffusion across an increasingly solidified polymer skin. The

lag behind the skin precipitation is dependent on the speed at which the polymer-rich phase

precipitates, which may range from near-instantaneous to a full minute. For good physical integrity

of the asymmetric membrane, it is critical that both phases of the solution remain above the critical

point of the phase diagram.19, 29

Figure 1-13. Depiction of the development of the polymer-rich and polymer-poor phases of the asymmetric membrane

during fabrication. 19, 29

While there are many variables impacting the structure of the resulting foam and the ASM

thickness, as well as the presence and abundance of voids in the foam, the overall procedure is

qualitatively well-represented by the above figures. With thorough empirical data quantitative

65

phase diagrams may also be constructed and used efficiently to optimize a phase inversion process.

Tie lines are established in the three-phase balance between polymer, solvent and nonsolvent. The

inclusion of additives in the nonsolvent bath is a frequent tactic to shift the point of precipitation,

which requires the construction of a second diagram or simple empirical study in order to

determine the optimum concentration for the desired impact on the phase separation process. 19, 29

1.7.5 ASM finishing techniques

The phase separation process is an effective way to produce a selective membrane structure

from a single material in a simple and efficient manner, although producing a defect-free

membrane is a challenge. To resolve any pinholes that may have formed, the membrane goes

through an annealing step post-production, by submerging into a high temperature bath, with the

exact conditions depending on the chemical structure of the membrane. This step causes small

pores and pinholes to collapse as the membrane approaches its hydrated glass transition

temperature, however it also causes thickening of the selective layer through the collapse of the

narrower end of the pore gradient, which is next to the skin layer.182, 183 This reduces the flux of

the membrane and so is applied carefully. 19, 29

In asymmetric membranes cast for gas separation, pinholes are sometimes resolved by the

introduction of what is termed a ‘gutter layer’ which is the application of a very thin layer of highly

permeable material over the selective layer of the ASM. While the gutter layer still permits the

easy and rapid passage of both the desired permeate and the components which are ideally

excluded, it does provide a barrier to the fluid flow behavior of the undesired components.

Likewise, it greatly improves rejection behavior in the event of one or several pinholes in the

membrane.

66

1.8 Sulfonated poly(arylene ether sulfone)s and membrane separations

1.8.1 Sulfonated poly(arylene ether sulfone)s

The poly(arylene ether) (PAES) family, typically with sulfonic acid derivatization, have

been explored in the literature as promising reverse osmosis membrane candidates, due to their

excellent mechanical properties and good chlorine resistance. Poly(arylene ether) copolymers are

encompass a range of related materials that have been investigated for possible use in reverse

osmosis and other membrane separations processes, including poly(arylene ether sulfones)s,

poly(arylene ether ketone)s, poly(arylene ether ether ketone)s, poly(ether ketone)s and other

copolymers. Nonsulfonated and sulfonated bisphenol-A and 4,4’-biphenol based poly(arylene

ether sulfone)s and related materials have been investigated in a variety of other membrane

applications, such as proton exchange membrane fuel cells (PEMFCs),184, 185 ultrafiltration,186-188

nanofiltration,189-191 and gas separation.192

Of the materials available in the poly(arylene ether) material family, PAES materials show

superior thermal stability due to the resonance structure afforded by the sulfone group and the high

oxidation state of the sulfur atom, which may allow for melt processability at up to 400oC in non-

ion-containing copolymers.193 The flexible ether linkages allow for reduced material rigidity, and

the mechanical properties and ion selectivity may be tailored by incorporating different functional

groups into the polymer backbone. In applications where partially sulfonated PAES copolymers

may offer advantages from both hydrophilic and hydrophobic phases, a synthetic approach

employing the synthesis of a multiblock copolymer or crosslinked copolymer may offer optimal

tailoring of membrane properties, i.e. improved salt rejection.194-196 While the adaptation of

multiblock partially sulfonated PAES copolymers to reverse osmosis applications is in the

67

relatively early stages, some generalizations may be made from the extensive research which has

been performed in the field of proton exchange membranes for fuel cells.

1.8.2 Post-sulfonated poly(arylene ether sulfone)s and reverse osmosis

The nonsulfonated poly(arylene ether sulfone)s (PAESs)have been used for some time as

the foam support in thin film composite membranes, as well as compaction-resistant porous

membranes for use in ultrafiltration and microfiltration. Research was performed in the 1980s on

post-sulfonated Udel ®, which was sulfonated after polymerization using fuming sulfuric acid or

chlorosulfonic acid, and its applicability as a selective membrane for reverse osmosis.169 The post-

sulfonated polysulfone was studied both as a dense membrane and as an asymmetric membrane,

although producing void-free and defect-free films was problematic.196

While the sulfonated polysulfones showed good potential, particularly for their tolerance

to chlorine exposure necessary for disinfection as well as for expected resistance to fouling, the

difficulty in tailoring the molecular weight and degree of sulfonation discouraged further research

for several decades. Asymmetric membranes cast from post-sulfonated polysulfones and

polyethersulfones have also been evaluated for use in nanofiltration and pervaporation, with some

promising performances but voids may form in the foam structure that detract from ASM

mechanical stability.197, 198

A possible alternative to the post-sulfonated PAES materials lies in polymers containing

small amounts of hydroquinone units. John Rose demonstrated that in poly(phenylene ether

sulfone)s, poly(phenylene ether ether sulfone)s and related polymers, monosulfonation of a

hydroquinone co-monomer could be obtained exclusively without modification of the rest of the

68

polymer chain by dissolving in and reacting with sulfuric acid. This modification may provide a

novel alternative material for reverse osmosis applications.199-202

1.8.3 Directly copolymerized sulfonated poly(arylene ether sulfone)s and

reverse osmosis

With the development in the early 2000s of sulfonated poly(arylene ether sulfone)s from

the direct copolymerization of the sulfonated monomer, 3,3’-disulfonated, 4,4’dichlorodiphenyl

sulfone (SDCDPS), 4,4’dichlorodiphenyl sulfone (DCDPS), and biphenol, a greater degree of

chemical stability as well as control and reproducibility of both sulfonation and molecular weight

prompted renewed interest in this material, particularly with the introduction of crosslinking.194

The synthesis of SDCDPS monomer used in the production of the directly copolymerized

poly(arylene ether sulfone)s is shown in Figure 1-14, and has not only been well documented in

the literature but also has well established procedures to confirm monomer purity which allows for

precise tailoring of the polymer structure.185, 203

69

Figure 1-14. Synthesis of SDCDPS used in the production of directly polymerized sulfonated PAESs.

The structures of post-sulfonated and directly copolymerized PAES are compared in Figure

1-15, where the greater stability of the sulfonate groups on the deactivated sulfone group should

be noted. Although the directly polymerized sulfonated PAES materials were initially developed

for use in proton exchange membrane fuel cells as proton conductors, the studies on water uptake

and transport properties provided a helpful foundation for reopening studies on applications for

reverse osmosis. A new synthetic approach has been developed to overcome the previous

disadvantages of an RO membrane, a series of biphenol- and bisphenol A-based copolymers were

synthesized and evaluated.10, 204-209

Figure 1-15. Comparison of post-sulfonated and directly copolymerized PAES structures.

Initial studies into the application of SPAES copolymers for reverse osmosis focused on

the structure-property relationships of these materials as a function of sulfonation level. Increases

in the hydrophilic group produced a marked increase in the water flux, but produced a sharp

70

decrease in salt rejection and a decrease in the mechanical properties of the hydrated membrane.

SPAES copolymers with a higher free volume tend to show higher water flux and slightly lower

salt rejection values due to the increased volume fraction and self-diffusion coefficient of water,

according to studies performed at Virginia Tech and the University of Texas.10, 204, 209, 210

Since the development of directly copolymerized SPAESs, investigations inside and

outside of the McGrath group have focused on improved salt rejection and water flux

characteristics. One approach is to crosslink a dense membrane in order to limit water uptake in

the membrane and fix hydrophilic groups into a more dense arrangement, thus improving salt

rejection.194, 211, 212 Another approach is to investigate the use of a multiblock copolymer

membrane, where the mechanical properties of a hydrophobic block and the transport properties

of a hydrophilic block might be optimized with nanophase-separated morphology.213

1.9 Characterization of RO Membranes

In the development of RO membranes, many factors beyond the raw material’s ability to

select for the transport of water while rejecting sodium chloride must be evaluated in the tuning of

a new system. Membranes must be hydrolytically stable over a range of pH levels for long

durations of time, and preferably should be resistant to degradation by common disinfectants and

other additives, such as chlorine. Resistance to fouling is also ideal, and selectivity for ions and

contaminants other than sodium and chloride, i.e. boron, arsenic, and pharmaceuticals would be

ideal. A membrane which is capable of withstanding the hydrocarbon contaminants in oily water

without degrading or experiencing a significant drop in rejection is also greatly desired and under

investigation.214 When producing an asymmetrical membrane or thin film composite, confirmation

of both the film and support foam layer structures is important both to observe continuous film

thickness, the absence of pore plugging in the foam, and the absence of pinholes in the skin layer.

71

While the final testing must be performed inside of a RO testing unit in order to understand

the salt rejection and water flux behavior, prior to RO evaluation a number of other evaluations

are performed in order to determine that a given material or membrane is a suitable RO candidate.

Mechanical property testing can help indicate whether a membrane can withstand the stress of the

applied pressure inside a module. Dynamic mechanical analysis (DMA) can help to identify low-

temperature transitions that may help a material to be more stress resilient, or in the case of a cross-

linked or multiblock copolymer, to identify the increase in glass transition temperature or the

degree of phase separation between blocks. AFM is useful to characterize a multi-component

surface such a nanophase-separated multiblock membrane, or to quantify the amount of surface

roughness in the case of a cross-linked polymer, which frequently has surface inconsistencies.

Imaging via scanning electron microscopy may help quantify skin thickness of an ASM or TFC,

as well as pore size and pore size gradients in the case of the former.

1.9.1 Mechanical properties

Tensile testing is frequently used to complement DMA evaluation. Instron testing is ideal for the

determination of mechanical properties such as modulus, yield stress, yield strain, and stress/strain

at break. This test is performed by punching out samples with a uniform die to insure minimal

variation in sample variations, typically using a dogbone shape. The dogbone shape is preferred

because stresses focus on the narrow portion of the film, away from the thicker ends used for

attaching clamps. During testing, the clamps are secured at each end and the sample is elongated

at a constant rate, while the force needed to maintain this elongation is measured.215

The stress is measured by the following equation, where represents stress, typically in

units of Pa, F is the applied load in units of N, and A is the initial cross-sectional area of the narrow

portion of the dogbone (m2).

72

𝜎 =𝐹

𝐴

The strain, , is determined by the change in length over the original length, and is unitless

but typically reported in percent (%).

휀 =∆𝐿

𝐿

𝐸 =𝜎

𝜀

The ratio of the stress to the strain is the Young’s Modulus, E, and their relationship

described in the above equation is known as Hooke’s Law. It is commonly used to gauge the

stiffness or rigidity of a material. The Young’s modulus is measured from the initial portion of a

stress-strain curve as shown in Figure 1-16, prior to the material’s yielding or fracture, where the

deformation is elastic or reversible. After the amount of elastic deformation possible in the sample

has been exhausted, the stress peaks with additional strain and then begins to decrease as

irreversible deformation occurs, known as the yield point. The sample then undergoes a phase of

elongation at a constant, lower stress called necking, where the thin portion of the dogbone

becomes much narrower until the sample reaches the point of failure.215

73

Figure 1-16. Model stress-strain curve of a polymer with elastic and nonelastic deformation.

1.9.2 Thermomechanical behavior via dynamic mechanical analysis

In dynamic mechanical analysis, a material’s transition temperature(s) are measured in terms of

the change of storage and loss moduli (E’ and E’’ respectively) as a function of temperature and/or

frequency. While DMA may be performed on thin films, fibers, or on bars in single or double

cantilever testing, all samples entail a material being fixed into a given frame and a deforming

oscillating force being applied at one end, and the response to that force is measured. When a force

is applied to a sample, there is a small lag before the material responds by deforming. The

difference between this initial applied stress, , and the material response, , is called the phase

lag, represented by . The storage modulus E’ and the loss modulus E” are determined by the

following equations:

𝐸′ =𝜎𝑜

𝜀𝑜cos 𝛿

74

𝐸" =𝜎𝑜

𝜀𝑜sin 𝛿

The most important information available in a DMA test is that of thermal transitions.

While other instrumentation can detect glass transition shifts, DMA is much more sensitive to the

onset of the glass transition. It can frequently also pick up beta and gamma transitions, which are

attributed to smaller molecular motions of side groups and bending and stretching within the chain.

These transitions are frequently missed in analysis using the less sensitive differential scanning

calorimetry (DSC). If a plasticizing agent has been added in small quantity, the greater sensitivity

of the DMA may be useful for confirming the Tg depression observed in DSC. In the case of a

blend, the Tg of the minority polymer can be detected as low as 2% content.

While the storage and loss moduli are also useful information, they are typically not the objective

of DMA testing. Young’s modulus is better measured from Instron testing, but an approximation

can be found for an estimate by determining the complex modulus from DMA, calculated by the

following equation:

𝐸∗ = 𝐸′ + 𝑖𝐸"

1.9.3 Scanning Electron Microscopy

Scanning Electron Microscopy, or SEM, is an imaging technique in which an electron

beam is used to bombard a sample. If the sample is electronically conductive and stable under ultra

high vacuum (UHV), it can be imaged without further processing. A sample which is not

electronically conductive must be sputter-coated with a conductive material such as gold or silver

prior to imaging, or painted with carbon. Samples which must be partially humidified, such as

biological samples, may be imaged in an environmental SEM (E-SEM), where a lower vacuum is

held at low temperature, albeit at reduced resolution. For reverse osmosis applications, SEM

75

imaging is critical for imaging not only pore size and pore gradients in the cross-section of a thin

film composite or asymmetric membrane, but also the selective skin thickness. It can also be used

for a qualitative analysis of the membrane surface roughness before proceeding with the more

time-consuming quantitative characterization possible with atomic force microscopy (AFM).29

1.9.4 Atomic Force Microscopy

Atomic Force Microscopy (AFM) is the primary application of scanning probe microscopy

(SPM), in which information about the height and surface hardness of a given material is obtained

by scanning a nanomachined probe tip with a defined rigidity across a small defined surface area.

This information is compiled, line by line, to create a composite ‘image’ which illustrates the

material morphology and topography. AFM is one of the few characterization techniques which

features atomic resolution, although most imaging is performed on a much coarser level. The

quality of imaging resolution and reproducibility is most dependent on the tip quality and

compatibility with the sample surface in terms of spring constant, applied force, and tip

composition, as well as a sample surface and working environment free of contamination.216

While scanning, the probe tip is maintained at a constant height above the material surface

by feedback to the piezoelectric actuator upon which it is mounted. This feedback is acquired from

a laser beam which bounces off of the probe tip as it scans across the sample surface, and then

lands upon a photodiode. The position of the beam on the photodiode is used to both calculate the

surface topography and to determine the amount of correction necessary to return the probe tip to

a constant height.216

During each scan of the probe tip, there are several different types of energetic interactions

between the tip and surface. At very low tip-surface distances favored by contact mode AFM,

76

repulsive forces dominate. As the tip moves further away from the surface, the repulsive forces

quickly drop off and are replaced by attractive forces such as Van der Waals, which is the region

typically favored by non-contact mode.216

There are three primary modes for collecting AFM images, non-contact mode, contact

mode (in which the scanning probe or tip comes into actual contact with the material surface) and

tapping mode (in which the probe or tip is repelled from the surface). There are also a number of

specialized applications of AFM, such as electrostatic force microscopy, chemical force

microscopy, AFM coupled with thermomechanical analysis, and magnetic force microscopy.216

Non-contact mode is distinguished by the distance at which the tip oscillates, typically 5-

15 nm from the surface, without physical contact with the surface at any point. At this distance,

while long range Van der Waals forces are detected by the tip, the repulsive and much of the

attractive forces which predominate tapping and contact mode are not influential. However, non-

contact mode does leave the tip susceptible to errors caused by any solid or liquid components

suspended in the air which have adsorbed onto the sample area.216

Contact mode is distinguished by the tip maintained in near-actual physical contact with or

effectively in contact with the sample surface, where repulsive atomic forces push the tip away.

Constant deflection maintains the distance between the surface and the tip static. Tapping mode is

a hybrid of both contact and non-contact mode in which the tip is lowered while osciallating at

resonance frequency until it comes into physical contact with the tip surface. The oscillations

prevent destruction of the fragile tip as the tip is lifted from the surface before any features on the

surface may damage it. This mode is ideal for the high-resolution topographical mapping of

complex surfaces.216

77

In the tapping mode, images are constructed depicting both the scale of surface rigidity

(phase image) and the physical topography of the imaged surface (height image). The information

from the latter can be understood qualitatively, from viewing the image, or quantitatively by

analyzing the height information collected during imaging to calculate the root mean square of the

heights, or RMS. This is a useful descriptor of the average surface roughness, so long as the image

captured is reflective of the rest of the membrane surface. This is particularly useful when

characterizing membranes for RO applications as surface roughness in a RO module creates dead

zones where low fluid turbulence enables foulants, particularly microbes, to settle and adhere to

the membrane surface. A lower RMS membrane fabricated from the same material as a high RMS

membrane could be expected to feature lower levels of fouling.216

References

1. Elimelech, M.; Phillip, W. A., The Future of Seawater Desalination: Energy, Technology, and the

Environment. Science 2011, 333 (6043), 712-717.

2. Wilf, M. A. L., The guidebook to membrane desalination technology : reverse osmosis, nanofiltration and hybrid systems : process, design, applications and economics. Balaban Desalination

Publications: L'Aquila, Italy, 2007.

3. Escobar, I. C. S. A. I., Sustainable water for the future : water recycling versus desalination.

Elsevier Science: Amsterdam; Boston, 2010.

4. Greenlee, L. F.; Lawler, D. F.; Freeman, B. D.; Marrot, B.; Moulin, P., Reverse osmosis

desalination: Water sources, technology, and today's challenges. Water Res. 2009, 43 (9), 2317-2348.

5. Hager, L. S. W. E. F., Membrane systems for wastewater treatment. WEF Press : McGraw-Hill:

New York, 2006.

6. Dawoud, M. A., The role of desalination in augmentation of water supply in GCC countries.

Desalination 2005, 186 (1-3), 187-198.

7. National Research Council, C. o. A. D. T. N. A. P., Desalination : a national perspective.

National Academies Press: Washington, D.C., 2008.

8. McDonald, R. I.; Green, P.; Balk, D.; Fekete, B. M.; Revenga, C.; Todd, M.; Montgomery, M.,

Urban growth, climate change, and freshwater availability. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (15),

6312-6317, S6312/1-S6312/2.

9. Thorne, O. M.; Fenner, R. A., Risk-based climate-change impact assessment for the water

industry. Water Sci Technol 2009, 59 (3), 443-51.

78

10. Geise, G. M.; Lee, H.-S.; Miller, D. J.; Freeman, B. D.; McGrath, J. E.; Paul, D. R., Water

purification by membranes: the role of polymer science. J. Polym. Sci., Part B: Polym. Phys. 2010, 48

(15), 1685-1718.

11. Tsiourtis, N. X., Desalination and the environment. Desalination 2001, 141 (3), 223-236.

12. Service, R. F., Desalination Freshens Up. Science 2006, 313 (5790), 1088-1090.

13. Khawaji, A. D.; Kutubkhanah, I. K.; Wie, J.-M., Advances in seawater desalination technologies.

Desalination 2008, 221 (1–3), 47-69.

14. Bergman, R., Reverse osmosis and nanofiltration. American Water Works Association: Denver,

CO, 2007.

15. Amjad, Z., Reverse osmosis : membrane technology, water chemistry & industrial applications.

Van Nostrand Reinhold: New York, 1993.

16. Kucera, J., Reverse osmosis : design, processes, and applications for engineers. Scrivener Pub. ;

Wiley: Salem, Mass.; Hoboken, N.J., 2010.

17. Focazio, M. J.; Kolpin, D. W.; Barnes, K. K.; Furlong, E. T.; Meyer, M. T.; Zaugg, S. D.; Barber,

L. B.; Thurman, M. E., A national reconnaissance for pharmaceuticals and other organic wastewater

contaminants in the United States - (II) Untreated drinking water sources. Sci. Total Environ. 2008, 402

(2-3), 201-216. 18. Focazio, M. J.; Kolpin, D. W.; Furlong, E. T. In Occurrence of human pharmaceuticals in water

resources of the United States: a review, Springer GmbH: 2004; pp 91-105.

19. Baker, R. W., Membrane Technology and Applications. 2004.

20. Li, N. N., Advanced membrane technology and applications. Wiley: Hoboken, N.J., 2008.

21. Malaeb, L.; Ayoub, G. M., Reverse osmosis technology for water treatment: State of the art

review. Desalination 2011, 267 (1), 1-8.

22. Penate, B.; Garcia-Rodriguez, L., Reverse osmosis hybrid membrane inter-stage design. A

comparative performance assessment. Desalination 2011, 281, 354-363.

23. Al-Sahali, M.; Ettouney, H., Developments in thermal desalination processes: Design, energy,

and costing aspects. Desalination 2007, 214 (1–3), 227-240.

24. Scenna, N. J., Synthesis of thermal desalination processes. Part I. Multistage flash distillation system (MSF). Desalination 1987, 64 (0), 111-122.

25. Reddy, K. V.; Ghaffour, N., Overview of the cost of desalinated water and costing

methodologies. Desalination 2007, 205 (1-3), 340-353.

26. Summers, E. K.; Arafat, H. A.; Lienhard, J. H. V., Energy efficiency comparison of single-stage

membrane distillation (MD) desalination cycles in different configurations. Desalination 2012, 290, 54-

66.

27. Almulla, A.; Hamad, A.; Gadalla, M., Integrating hybrid systems with existing thermal

desalination plants. Desalination 2005, 174 (2), 171-192.

28. Prakash, S.; Bellman, K.; Shannon, M. A. In Recent advances in water desalination through

biotechnology and nanotechnology, CRC Press: 2012; pp 365-382.

29. Wang, L. K., Membrane and Desalination Technologies. Springer: 2008.

30. Van, d. B. B.; Vandecasteele, C., Distillation vs. membrane filtration: overview of process

evolutions in seawater desalination. Desalination 2002, 143 (3), 207-218.

31. Charcosset, C., A review of membrane processes and renewable energies for desalination.

Desalination 2009, 245 (1-3), 214-231.

32. Garcia-Rodriguez, L., Renewable energy applications in desalination: state of the art. Sol. Energy 2003, 75 (5), 381-393.

33. Alklaibi, A. M.; Lior, N., Membrane-distillation desalination: Status and potential. Desalination

2005, 171 (2), 111-131.

34. Al-Obaidani, S.; Curcio, E.; Macedonio, F.; Di Profio, G.; Al-Hinai, H.; Drioli, E., Potential of

membrane distillation in seawater desalination: Thermal efficiency, sensitivity study and cost estimation.

Journal of Membrane Science 2008, 323 (1), 85-98.

79

35. Qtaishat, M. R.; Banat, F., Desalination by solar powered membrane distillation systems.

Desalination 2013, 308 (0), 186-197.

36. Susanto, H., Towards practical implementations of membrane distillation. Chemical Engineering and Processing: Process Intensification 2011, 50 (2), 139-150.

37. Nielsen, W., Membrane Filtration and Related Molecular Separation Technologies. APV

Systems: 2000.

38. Bonnelye, V.; Sanz, M. A.; Durand, J.-P.; Plasse, L.; Gueguen, F.; Mazounie, P., Reverse

osmosis on open intake seawater: Pre-treatment strategy. Desalination 2004, 167 (1-3), 191-200.

39. Brehant, A.; Bonnelye, V.; Perez, M., Assessment of ultrafiltration as a pretreatment of reverse

osmosis membranes for surface seawater desalination. Water Sci. Technol.: Water Supply 2003, 3 (5-6),

437-445.

40. American Water Works, A., Microfiltration and ultrafiltration membranes for drinking water.

American Water Works Association: Denver, CO, 2005.

41. Kang, S.-T.; Subramani, A.; Hoek, E. M. V.; Deshusses, M. A.; Matsumoto, M. R., Direct

observation of biofouling in cross-flow microfiltration: mechanisms of deposition and release. J. Membr.

Sci. 2004, 244 (1-2), 151-165.

42. Lebleu, N.; Roques, C.; Aimar, P.; Causserand, C., Role of the cell-wall structure in the retention of bacteria by microfiltration membranes. J. Membr. Sci. 2009, 326 (1), 178-185.

43. Afonso, M. D.; Jaber, J. O.; Mohsen, M. S., Brackish groundwater treatment by reverse osmosis

in Jordan. Desalination 2004, 164 (2), 157-171.

44. Tu, K. L.; Nghiem, L. D.; Chivas, A. R., Boron removal by reverse osmosis membranes in

seawater desalination applications. Sep. Purif. Technol. 2010, 75 (2), 87-101.

45. Tu, K.-L.; Chivas, A. R.; Nghiem, L. D., Effects of membrane fouling and scaling on boron

rejection by nanofiltration and reverse osmosis membranes. Desalination 2011, 279 (1-3), 269-277.

46. Mondal, S.; Hsiao, C.-I.; Wickramasinghe, S. R., Nanofiltration/reverse osmosis for treatment of

coproduced waters. Environ. Prog. 2008, 27 (2), 173-179.

47. Mondal, S.; Wickramasinghe, S. R., Produced water treatment by nanofiltration and reverse

osmosis membranes. Journal of Membrane Science 2008, 322 (1), 162-170. 48. Amiri, M. C.; Samiei, M., Enhancing permeate flux in a RO plant by controlling membrane

fouling. Desalination 2007, 207 (1-3), 361-369.

49. Lee, K. P.; Arnot, T. C.; Mattia, D., A review of reverse osmosis membrane materials for

desalination. Development to date and future potential. J. Membr. Sci. 2011, 370 (1-2), 1-22.

50. Mathioulakis, E.; Belessiotis, V.; Delyannis, E., Desalination by using alternative energy: Review

and state-of-the-art. Desalination 2007, 203 (1-3), 346-365.

51. Mezher, T.; Fath, H.; Abbas, Z.; Khaled, A., Techno-economic assessment and environmental

impacts of desalination technologies. Desalination 2011, 266 (1-3), 263-273.

52. Karagiannis, I. C.; Soldatos, P. G., Water desalination cost literature: review and assessment.

Desalination 2008, 223 (1-3), 448-456.

53. Klaysom, C.; Cath, T. Y.; Depuydt, T.; Vankelecom, I. F. J., Forward and pressure retarded

osmosis: potential solutions for global challenges in energy and water supply. Chem. Soc. Rev. 2013, 42

(16), 6959-6989.

54. Achilli, A.; Childress, A. E., Pressure retarded osmosis: From the vision of Sidney Loeb to the

first prototype installation - Review. Desalination 2010, 261 (3), 205-211.

55. Loeb, S., Large-scale power production by pressure-retarded osmosis, using river water and sea

water passing through spiral modules. Desalination 2002, 150 (2), 205.

56. Loeb, S., One hundred and thirty benign and renewable megawatts from Great Salt Lake? The

possibilities of hydroelectric power by pressure-retarded osmosis with spiral module membranes:

Desalination, 141 (2001) 85–91. Desalination 2002, 142 (2), 207.

57. Cath, T. Y.; Childress, A. E.; Elimelech, M., Forward osmosis: Principles, applications, and

recent developments. Journal of Membrane Science 2006, 281 (1–2), 70-87.

80

58. Jin, X.; She, Q.; Ang, X.; Tang, C. Y., Removal of boron and arsenic by forward osmosis

membrane: Influence of membrane orientation and organic fouling. J. Membr. Sci. 2012, 389, 182-187.

59. Xie, M.; Nghiem, L. D.; Price, W. E.; Elimelech, M., Comparison of the removal of hydrophobic

trace organic contaminants by forward osmosis and reverse osmosis. Water Res. 2012, 46 (8), 2683-2692.

60. Duong, P. H. H.; Zuo, J.; Chung, T.-S., Highly crosslinked layer-by-layer polyelectrolyte FO

membranes: Understanding effects of salt concentration and deposition time on FO performance. J.

Membr. Sci. 2013, 427, 411-421.

61. Huang, L.; Bui, N.-N.; Meyering, M. T.; Hamlin, T. J.; McCutcheon, J. R., Novel hydrophilic

nylon 6,6 microfiltration membrane supported thin film composite membranes for engineered osmosis.


62. Arena, J. T.; McCloskey, B.; Freeman, B. D.; McCutcheon, J. R., Surface modification of thin

film composite membrane support layers with polydopamine: Enabling use of reverse osmosis

membranes in pressure retarded osmosis. Journal of Membrane Science 2011, 375 (1–2), 55-62.

63. Zhang, S.; Fu, F.; Chung, T.-S., Substrate modifications and alcohol treatment on thin film

composite membranes for osmotic power. Chemical Engineering Science 2013, 87 (0), 40-50.

64. Alsvik, I. L.; Hagg, M.-B., Preparation of thin film composite membranes with polyamide film on

hydrophilic supports. J. Membr. Sci. 2013, 428, 225-231. 65. Han, G.; Zhang, S.; Li, X.; Widjojo, N.; Chung, T.-S., Thin film composite forward osmosis

membranes based on polydopamine modified polysulfone substrates with enhancements in both water

flux and salt rejection. Chem. Eng. Sci. 2012, 80, 219-231.

66. Hoover, L. A.; Schiffman, J. D.; Elimelech, M., Nanofibers in thin-film composite membrane

support layers: Enabling expanded application of forward and pressure retarded osmosis. Desalination

2013, 308, 73-81.

67. Jeong, B.-R.; Kim, J.-H.; Kim, B.-S.; Park, Y.-I.; Song, D.-H.; Kim, I.-C., Effect of support

membrane property on performance of forward osmosis membrane. Membr. J. 2010, 20 (3), 235-240.

68. Lee, S. H.; Yoo, Y. B.; Seo, S. G. Method for manufacturing forward osmosis membrane for

water treatment. KR990168B1, 2010.

69. Subramani, A.; Badruzzaman, M.; Oppenheimer, J.; Jacangelo, J. G., Energy minimization strategies and renewable energy utilization for desalination: A review. Water Res. 2011, 45 (5), 1907-

1920.

70. Wang, R.; Shi, L.; Tang, C. Y.; Chou, S.; Qiu, C.; Fane, A. G., Characterization of novel forward

osmosis hollow fiber membranes. J. Membr. Sci. 2010, 355 (1-2), 158-167.

71. Thorsen, T.; Holt, T., The potential for power production from salinity gradients by pressure

retarded osmosis. J. Membr. Sci. 2009, 335 (1+2), 103-110.

72. Thorsen, T.; Holt, T., Finding hidden energy in membrane processes. Filtr. Sep. 2005, 42 (3), 28-

30.

73. Cath, T. Y., Osmotically and thermally driven membrane processes for enhancement of water

recovery in desalination processes. Desalin. Water Treat. 2010, 15 (1-3), 279-286.

74. Achilli, A.; Cath, T. Y.; Childress, A. E. In Inorganic draw solutions for forward osmosis

processes, American Chemical Society: 2010; pp ENVR-69.

75. Ge, Q.; Ling, M.; Chung, T.-S., Draw solutions for forward osmosis processes: Developments,

challenges, and prospects for the future. J. Membr. Sci. 2013, 442, 225-237.

76. Ge, Q.; Su, J.; Amy, G. L.; Chung, T.-S., Exploration of polyelectrolytes as draw solutes in

forward osmosis processes. Water Res. 2012, 46 (4), 1318-1326.

77. Stewart, F. F.; Benson, M. T.; Stone, M. L. Draw solutes, methods of forming draw solutes, and

methods of using draw solutes to treat an aqueous liquid. US20130048564A1, 2013.

78. Stone, M. L.; Rae, C.; Stewart, F. F.; Wilson, A. D., Switchable polarity solvents as draw solutes

for forward osmosis. Desalination 2013, 312, 124-129.

79. Stone, M. L.; Wilson, A. D.; Harrup, M. K.; Stewart, F. F., An initial study of hexavalent

phosphazene salts as draw solutes in forward osmosis. Desalination 2013, 312, 130-136.

81

80. Achilli, A.; Cath, T. Y.; Childress, A. E., Power generation with pressure retarded osmosis: An

experimental and theoretical investigation. J. Membr. Sci. 2009, 343 (1-2), 42-52.

81. McCutcheon, J. R.; Elimelech, M., Influence of concentrative and dilutive internal concentration

polarization on flux behavior in forward osmosis. J. Membr. Sci. 2006, 284 (1+2), 237-247.

82. Chung, T.-S.; Zhang, S.; Wang, K. Y.; Su, J.; Ling, M. M., Forward osmosis processes:

Yesterday, today and tomorrow. Desalination 2012, 287, 78-81.

83. Li, G.; Li, X.; Wang, D.; He, T.; Gao, C., Forward osmosis membranes and applications.

Huagong Jinzhan 2010, 29 (8), 1388-1398.

84. McCutcheon, J.; Bui, N.-N. In Forward osmosis, Scrivener Publishing LLC: 2014; pp 255-285.

85. Zhao, S.; Zou, L.; Tang, C. Y.; Mulcahy, D., Recent developments in forward osmosis:

Opportunities and challenges. J. Membr. Sci. 2012, 396, 1-21.

86. Wang, K. Y.; Ong, R. C.; Chung, T.-S., Double-Skinned Forward Osmosis Membranes for

Reducing Internal Concentration Polarization within the Porous Sublayer. Ind. Eng. Chem. Res. 2010, 49

(10), 4824-4831.

87. Widjojo, N.; Chung, T.-S.; Weber, M.; Maletzko, C.; Warzelhan, V., A sulfonated

polyphenylenesulfone (sPPSU) as the supporting substrate in thin film composite (TFC) membranes with

enhanced performance for forward osmosis (FO). Chemical Engineering Journal 2013, 220 (0), 15-23. 88. Yang, Q.; Wang, K. Y.; Chung, T.-S. Dual-layer hollow fibers with enhanced flux as forward

osmosis membranes for water reuses and protein enrichment. WO2010045430A2, 2010.

89. Field, R. W.; Wu, J. J., Mass transfer limitations in forward osmosis: Are some potential

applications overhyped? Desalination 2013, 318, 118-124.

90. Field, R. W.; Wu, J. J., Analysis of Forward Osmosis: Is it Overhyped? Procedia Eng. 2012, 44,

264-266.

91. Boo, C.; Elimelech, M.; Hong, S., Fouling control in a forward osmosis process integrating

seawater desalination and wastewater reclamation. J. Membr. Sci. 2013, 444, 148-156.

92. Post, J. W.; Veerman, J.; Hamelers, H. V. M.; Euverink, G. J. W.; Metz, S. J.; Nymeijer, K.;

Buisman, C. J. N., Salinity-gradient power: Evaluation of pressure-retarded osmosis and reverse

electrodialysis. J. Membr. Sci. 2007, 288 (1+2), 218-230. 93. Bowden, K. S.; Achilli, A.; Childress, A. E., Organic ionic salt draw solutions for osmotic

membrane bioreactors. Bioresour. Technol. 2012, 122, 207-216.

94. Wang, W.; Sun, J.; Wang, B., Effect of process parameters on osmosis performance during

forward osmosis. Gaofenzi Cailiao Kexue Yu Gongcheng 2012, 28 (1), 86-88,92.

95. Chung, T.-S.; Li, X.; Ong, R. C.; Ge, Q.; Wang, H.; Han, G., Emerging forward osmosis (FO)

technologies and challenges ahead for clean water and clean energy applications. Curr. Opin. Chem. Eng.

2012, 1 (3), 246-257.

96. Baker, R. W., Separation of volatile organic compounds from water by pervaporation. MRS Bull.

1999, 24 (3), 50-53.

97. Neel, J. In The pervaporation process, Oxford & IBH: 1992; pp 313-29.

98. Wijmans, J. G.; Baker, R. W.; Mairal, A. P. Membrane separation of organic mixtures using gas

separation or pervaporation and dephlegmation. US20030233934A1, 2003.

99. Zhu, C.; Xu, W.; Feng, J. In Description of pervaporation (PVAP) mechanism with separation-characteristics graphs, Bakish Mater. Corp.: 1988; pp 44-53.

100. Paul, D. R., Reformulation of the solution-diffusion theory of reverse osmosis. J. Membr. Sci.

2004, 241 (2), 371-386.

101. Paul, D. R., The solution-diffusion model for swollen membranes. Sep. Purif. Methods 1976, 5

(1), 33-50.

102. Wijmans, J. G.; Baker, R. W. In The solution-diffusion model: a unified approach to membrane permeation, John Wiley & Sons Ltd.: 2006; pp 159-189.

103. Edzwald, J. K.; Haarhoff, J., Seawater pretreatment for reverse osmosis: Chemistry,

contaminants, and coagulation. Water Res. 2011, 45 (17), 5428-5440.

82

104. Elguera, A. M.; Perez, B. S. O., Development of the most adequate pre-treatment for high

capacity seawater desalination plants with open intake. Desalination 2005, 184 (1-3), 173-183.

105. Kim, S.-H.; Min, C.-S.; Cho, J., Comparison of different pretreatments for seawater desalination.

Desalin. Water Treat. 2011, 32 (1-3), 339-344.

106. Al-Juboori, R. A.; Yusaf, T., Biofouling in RO system: Mechanisms, monitoring and controlling.

Desalination 2012, 302 (0), 1-23.

107. Flemming, H.-C., Reverse osmosis membrane biofouling. Experimental Thermal and Fluid Science 1997, 14 (4), 382-391.

108. Herzberg, M.; Elimelech, M. In The role of EPS in biofouling of reverse osmosis membranes,

American Chemical Society: 2007; pp SUST-088.

109. Herzberg, M.; Elimelech, M., Biofouling of reverse osmosis membranes: Mechanisms and

performance. Prepr. Ext. Abstr. ACS Natl. Meet., Am. Chem. Soc., Div. Environ. Chem. 2006, 46 (2),

1272-1275.

110. Bernstein, R.; Belfer, S.; Freger, V., Toward Improved Boron Removal in RO by Membrane

Modification: Feasibility and Challenges. Environ. Sci. Technol. 2011, 45 (8), 3613-3620.

111. Glueckstern, P.; Priel, M., Boron removal in brackish water desalination systems. Desalination

2007, 205 (1-3), 178-184. 112. Glueckstern, P.; Priel, M., Optimization of boron removal in old and new SWRO systems.

Desalination 2003, 156 (1-3), 219-228.

113. Jacob, C., Seawater desalination: boron removal by ion exchange technology. Desalination 2007,

205 (1-3), 47-52.

114. Boo, C.; Lee, S.; Elimelech, M.; Meng, Z.; Hong, S., Colloidal fouling in forward osmosis: Role

of reverse salt diffusion. J. Membr. Sci. 2012, 390-391, 277-284.

115. Hyun, J.; Jang, H.; Kim, K.; Na, K.; Tak, T., Restriction of biofouling in membrane filtration

using a brush-like polymer containing oligoethylene glycol side chains. J. Membr. Sci. 2006, 282 (1+2),

52-59.

116. Louie, J. S.; Pinnau, I.; Ciobanu, I.; Ishida, K. P.; Ng, A.; Reinhard, M., Effects of polyether-

polyamide block copolymer coating on performance and fouling of reverse osmosis membranes. J. Membr. Sci. 2006, 280 (1+2), 762-770.

117. Matin, A.; Khan, Z.; Zaidi, S. M. J.; Boyce, M. C., Biofouling in reverse osmosis membranes for

seawater desalination: Phenomena and prevention. Desalination 2011, 281, 1-16.

118. Mohammadi, T.; Esmaeelifar, A., Wastewater treatment of a vegetable oil factory by a hybrid

ultrafiltration-activated carbon process. J. Membr. Sci. 2005, 254 (1-2), 129-137.

119. Pandey, S. R.; Jegatheesan, V.; Baskaran, K.; Shu, L., Fouling in reverse osmosis (RO)

membrane in water recovery from secondary effluent: a review. Rev. Environ. Sci. Bio/Technol. 2012, 11

(2), 125-145.

120. Yang, H. L.; Huang, C.; Pan, J. R., Characteristics of RO foulants in a brackish water desalination

plant. Desalination 2008, 220 (1-3), 353-358.

121. Yiantsios, S. G.; Sioutopoulos, D.; Karabelas, A. J., Colloidal fouling of RO membranes: an

overview of key issues and efforts to develop improved prediction techniques. Desalination 2005, 183 (1-

3), 257-272.

122. Ng, H. Y.; Elimelech, M., Influence of colloidal fouling on rejection of trace organic

contaminants by reverse osmosis. J. Membr. Sci. 2004, 244 (1-2), 215-226.

123. Amjad, Z.; Pugh, J.; Harn, J., Membranes. Antiscalants and dispersants in reverse osmosis

systems. Ultrapure Water 1996, 13 (8), 48-52.

124. Malekar, S., Role of antiscalants in reverse osmosis. Chem. Ind. Dig. 2005, 18 (3), 47-48,51-52.

125. Abejon, R.; Garea, A.; Irabien, A., Analysis, modelling and simulation of hydrogen peroxide

ultrapurification by multistage reverse osmosis. Chem. Eng. Res. Des. 2012, 90 (3), 442-452.

126. Antony, A.; Low, J. H.; Gray, S.; Childress, A. E.; Le-Clech, P.; Leslie, G., Scale formation and

control in high pressure membrane water treatment systems: A review. Journal of Membrane Science

2011, 383 (1–2), 1-16.

83

127. Badruzzaman, M.; Subramani, A.; DeCarolis, J.; Pearce, W.; Jacangelo, J. G., Impacts of silica on

the sustainable productivity of reverse osmosis membranes treating low-salinity brackish groundwater.

Desalination 2011, 279 (1-3), 210-218.

128. Boffardi, B. P. In Scale and deposit control for reverse osmosis systems, American Water Works

Association: 1997; pp 681-693.

129. Bonne, P. A. C.; Hofman, J. A. M. H.; Van, d. H. J. P., Scaling control of RO membranes and

direct treatment of surface water. Desalination 2000, 132 (1-3), 109-119.

130. Lee, S.; Cho, J.; Elimelech, M., Influence of colloidal fouling and feed water recovery on salt

rejection of RO and NF membranes. Desalination 2004, 160 (1), 1-12.

131. Al-Amoudi, A. S., Factors affecting natural organic matter (NOM) and scaling fouling in NF

membranes: A review. Desalination 2010, 259 (1–3), 1-10.

132. Kim, S.; Park, N.; Lee, S.; Cho, J., Membrane characterizations for mitigation of organic fouling

during desalination and wastewater reclamation. Desalination 2009, 238 (1-3), 70-77.

133. Lee, S.; Ang, W. S.; Elimelech, M. In Role of divalent cations in organic fouling of reverse osmosis membranes, American Chemical Society: 2004; pp ENVR-264.

134. Lee, S.; Elimelech, M., Relating Organic Fouling of Reverse Osmosis Membranes to

Intermolecular Adhesion Forces. Environ. Sci. Technol. 2006, 40 (3), 980-987. 135. Misdan, N.; Lau, W. J.; Ismail, A. F., Seawater reverse osmosis (SWRO) desalination by thin-

film composite membrane: Current development, challenges, and future prospects. Desalination 2012,

287, 228-237.

136. Bereschenko, L. A.; Prummel, H.; Euverink, G. J. W.; Stams, A. J. M.; van, L. M. C. M., Effect

of conventional chemical treatment on the microbial population in a biofouling layer of reverse osmosis

systems. Water Res. 2011, 45 (2), 405-416.

137. Mansouri, J.; Harrisson, S.; Chen, V., Strategies for controlling biofouling in membrane filtration

systems: challenges and opportunities. J. Mater. Chem. 2010, 20 (22), 4567-4586.

138. Matsuura, T.; Ismail, A. F.; Rana, D. In Applications of surface modifying macromolecules for

various membrane separation processes, American Chemical Society: 2011; pp I+EC-112.

139. Xu, J.; Wang, Z.; Yu, L.; Wang, J.; Wang, S., Reverse osmosis membrane with regenerable anti -biofouling and chlorine resistant properties. J. Membr. Sci. 2013, 435, 80-91.

140. Sagle, A. C.; Van Wagner, E. M.; Ju, H.; McCloskey, B. D.; Freeman, B. D.; Sharma, M. M.,

PEG-coated reverse osmosis membranes: Desalination properties and fouling resistance. J. Membr. Sci.

2009, 340 (1-2), 92-108.

141. McCloskey, B. D.; Park, H. B.; Ju, H.; Rowe, B. W.; Miller, D. J.; Chun, B. J.; Kin, K.; Freeman,

B. D., Influence of polydopamine deposition conditions on pure water flux and foulant adhesion

resistance of reverse osmosis, ultrafiltration, and microfiltration membranes. Polymer 2010, 51 (15),

3472-3485.

142. Kang, G.-d.; Cao, Y.-m., Development of antifouling reverse osmosis membranes for water

treatment: A review. Water Res. 2012, 46 (3), 584-600.

143. Yip, N. Y.; Phillip, W. A.; Schiffman, J. D.; Elimelech, M. High flux thin-film composite forward

osmosis and pressure-retarded osmosis membranes. WO2011069050A1, 2011.

144. Hilal, N.; Kim, G. J.; Somerfield, C., Boron removal from saline water. A comprehensive review.

Desalination 2011, 273 (1), 23-35.

145. Mondal, P.; Majumder, C. B.; Mohanty, B., Laboratory based approaches for arsenic remediation

from contaminated water: Recent developments. J. Hazard. Mater. 2006, 137 (1), 464-479.

146. Galvin, R. M., Occurrence of metals in waters: An overview. Water SA 1996, 22 (1), 7-18.

147. Akin, I.; Arslan, G.; Tor, A.; Cengeloglu, Y.; Ersoz, M., Removal of arsenate [As(V)] and

arsenite [As(III)] from water by SWHR and BW-30 reverse osmosis. Desalination 2011, 281, 88-92.

148. Coronell, O.; Mi, B.; Marinas, B. J.; Cahill, D. G., Modeling the Effect of Charge Density in the

Active Layers of Reverse Osmosis and Nanofiltration Membranes on the Rejection of Arsenic(III) and

Potassium Iodide. Environ. Sci. Technol. 2013, 47 (1), 420-428.

84

149. Xu, P.; Capito, M.; Cath, T. Y., Selective removal of arsenic and monovalent ions from brackish

water reverse osmosis concentrate. J. Hazard. Mater. 2013, 260, 885-891.

150. Yoon, J.; Amy, G.; Chung, J.; Sohn, J.; Yoon, Y., Removal of toxic ions (chromate, arsenate, and

perchlorate) using reverse osmosis, nanofiltration, and ultrafiltration membranes. Chemosphere 2009, 77

(2), 228-235.

151. Yoon, J.; Amy, G.; Yoon, Y., Transport of target anions, chromate (Cr (VI)), arsenate (As (V)),

and perchlorate (ClO-4), through RO, NF, and UF membranes. Water Sci. Technol. 2005, 51 (6-7, Water

Environment--Membrane Technology), 327-334.

152. Xu, J.; Gao, X.; Chen, G.; Zou, L.; Gao, C., High performance boron removal from seawater by

two-pass SWRO system with different membranes. Water Sci. Technol.: Water Supply 2010, 10 (3), 327-

336.

153. Dydo, P.; Nems, I.; Turek, M., Boron removal and its concentration by reverse osmosis in the

presence of polyol compounds. Sep. Purif. Technol. 2012, 89, 171-180.

154. Blank, J. E.; Tusel, G. F.; Nisanc, S., The real cost of desalted water and how to reduce it further.

Desalination 2007, 205 (1-3), 298-311.

155. Bouguecha, S.; Hamrouni, B.; Dhahbi, M., Small scale desalination pilots powered by renewable

energy sources: case studies. Desalination 2005, 183 (1-3), 151-165. 156. Ghaffour, N.; Reddy, V. K.; Abu-Arabi, M., Technology development and application of solar

energy in desalination: MEDRC contribution. Renewable Sustainable Energy Rev. 2011, 15 (9), 4410-

4415.

157. Qiu, T. Y.; Davies, P. A., The scope to improve the efficiency of solar-powered reverse osmosis.

Desalin. Water Treat. 2011, 35 (1-3), 14-32.

158. Charcosset, C.; Falconet, C.; Combe, M., Hydrostatic pressure plants for desalination via reverse

osmosis. Renewable Energy 2009, 34 (12), 2878-2882.

159. Afrasiabi, N.; Shahbazali, E., RO brine treatment and disposal methods. Desalin. Water Treat. 2011, 35 (1-3), 39-53.

160. Melian-Martel, N.; Sadhwani, J. J.; Baez, S. O. P., Saline waste disposal reuse for desalination

plants for the chlor-alkali industry. The particular case of Poso Izquierdo SWRO desalination plant. Desalination 2011, 281, 35-41.

161. Voutchkov, N., Overview of seawater concentrate disposal alternatives. Desalination 2011, 273

(1), 205-219.

162. Loeb, S., The Loeb-Sourirajan membrane: how it came about. ACS Symp. Ser. 1981, 153 (Synth.

Membr., Vol. 1), 1-9.

163. Kang, G.-D.; Gao, C.-J.; Chen, W.-D.; Jie, X.-M.; Cao, Y.-M.; Yuan, Q., Study on hypochlorite

degradation of aromatic polyamide reverse osmosis membrane. J. Membr. Sci. 2007, 300 (1+2), 165-171.

164. Avlonitis, S.; Hanbury, W. T.; Hodgkiess, T., Chlorine degradation of aromatic polyamides.

Desalination 1992, 85 (3), 321-34.

165. Shintani, T.; Matsuyama, H.; Kurata, N., Development of a chlorine-resistant polyamide reverse

osmosis membrane. Desalination 2007, 207 (1-3), 340-348.

166. Liu, L.-F.; Yu, S.-C.; Wu, L.-G.; Gao, C.-J., Study on a novel antifouling polyamide-urea reverse

osmosis composite membrane (ICIC-MPD). III. Analysis of membrane electrical properties. J. Membr. Sci. 2008, 310 (1+2), 119-128.

167. Liu, L.-F.; Xu, D.-Z.; Mao, P.-Q.; Zhang, L.; Gao, C.-J., Preparation and characterization of a

novel polyimide-urethane reverse osmosis composite membrane material. Gaodeng Xuexiao Huaxue Xuebao 2012, 33 (7), 1605-1612.

168. McGrath, J. E.; Park, H. B.; Freeman, B. D. Chlorine resistant desalination membranes based on

directly sulfonated poly(arylene ether sulfone) copolymers. US20070163951A1, 2007.

169. McGrath, J. E.; Wightman, J. P.; Lloyd, D. R. Novel poly(aryl ether) membranes for desalination by reverse osmosis; Virginia Polytech. Inst. and State Univ.: 1984; p 139 pp.

170. !!! INVALID CITATION !!! {}.

85

171. Kang, G.-d.; Liu, Z.-n.; Yu, H.-j.; Cao, Y.-m., Enhancing antifouling property of commercial

polyamide reverse osmosis membrane by surface coating using a brush-like polymer containing poly

(ethylene glycol) chains. Desalin. Water Treat. 2012, 37 (1-3), 139-145.

172. Yu, S.; Lue, Z.; Chen, Z.; Liu, X.; Liu, M.; Gao, C., Surface modification of thin-film composite

polyamide reverse osmosis membranes by coating N-isopropylacrylamide-co-acrylic acid copolymers for

improved membrane properties. J. Membr. Sci. 2011, 371 (1-2), 293-306.

173. Yip, N. Y.; Tiraferri, A.; Phillip, W. A.; Schiffman, J. D.; Elimelech, M., High Performance

Thin-Film Composite Forward Osmosis Membrane. Environ. Sci. Technol. 2010, 44 (10), 3812-3818.

174. Alsvik, I. L.; Hagg, M.-B., Pressure retarded osmosis and forward osmosis membranes: materials

and methods. Polymers (Basel, Switz.) 2013, 5 (1), 303-327, 25 pp.

175. Wang, K. Y.; Yang, Q.; Chung, T.-S.; Rajagopalan, R., Enhanced forward osmosis from

chemically modified polybenzimidazole (PBI) nanofiltration hollow fiber membranes with a thin wall.

Chem. Eng. Sci. 2009, 64 (7), 1577-1584.

176. Bui, N.-N.; McCutcheon, J. R., Hydrophilic Nanofibers as New Supports for Thin Film

Composite Membranes for Engineered Osmosis. Environ. Sci. Technol. 2013, 47 (3), 1761-1769.

177. Jeong, B.-H.; Hoek, E. M. V.; Yan, Y.; Subramani, A.; Huang, X.; Hurwitz, G.; Ghosh, A. K.;

Jawor, A., Interfacial polymerization of thin film nanocomposites: A new concept for reverse osmosis membranes. J. Membr. Sci. 2007, 294 (1+2), 1-7.

178. Widjojo, N.; Chung, T.-S.; Weber, M.; Maletzko, C.; Warzelhan, V., The role of sulphonated

polymer and macrovoid-free structure in the support layer for thin-film composite (TFC) forward osmosis

(FO) membranes [Erratum to document cited in CA155:616441]. J. Membr. Sci. 2012, 389, 544.

179. Sourirajan, S. M. T. A. C. S. D. o. I.; Engineering Chemistry, A. C. S. M. In Reverse osmosis and

ultrafiltration, Washington, D.C., 1985; American Chemical Society: Washington, D.C.

180. Pinnau, I.; Koros, W. J., A qualitative skin-layer formation mechanism for membranes made by

dry/wet phase inversion. J. Polym. Sci., Part B: Polym. Phys. 1993, 31 (4), 419-27.

181. So, M. T.; Eirich, F. R.; Strathmann, H.; Baker, R. W., Preparation of asymmetric Loeb-

Sourirajan membranes. J. Polym. Sci., Polym. Lett. Ed. 1973, 11 (3), 201-5.

182. Duarte, A. P.; Bordado, J. C.; Cidade, M. T., Cellulose acetate reverse osmosis membranes: optimization of preparation parameters. J. Appl. Polym. Sci. 2007, 103 (1), 134-139.

183. Su, J.; Zhang, S.; Chen, H.; Chen, H.; Jean, Y. C.; Chung, T.-S., Effects of annealing on the

microstructure and performance of cellulose acetate membranes for pressure-retarded osmosis processes.

J. Membr. Sci. 2010, 364 (1-2), 344-353.

184. Bae, B.; Miyatake, K.; Watanabe, M., Sulfonated Poly(arylene ether sulfone ketone) Multiblock

Copolymers with Highly Sulfonated Block. Synthesis and Properties. Macromolecules (Washington, DC,

U. S.) 2010, 43 (6), 2684-2691.

185. Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath, J. E., Alternative Polymer

Systems for Proton Exchange Membranes (PEMs). Chem. Rev. (Washington, DC, U. S.) 2004, 104 (10),

4587-4611.

186. Bosak, V. Z.; Vakulyuk, P. V.; Burban, A. F.; Isaev, S. D.; Vortman, M. Y.; Klimenko, N. S.;

Shevchenko, V. V., Charged ultrafiltration polysulfone membranes formed in the presence of ionogenic

surface-active oligomers. Dopov. Nats. Akad. Nauk Ukr. 2007, (11), 130-134.

187. Bosak, V. Z.; Vakulyuk, P. V.; Burban, A. F.; Antonyuk, N. G.; Isaev, S. D.; Lavrik, V. I.,

Production of charged polysulfone ultrafiltration membranes and the study of their properties. Dopov.

Nats. Akad. Nauk Ukr. 2007, (8), 127-132.

188. Bosak, V. Z.; Vakulyuk, P. V.; Burban, A. F., Surface modification of polysulfone membranes by

UV-initiated graft polymerization of N-vinyl-2-pyrrolidone. Ukr. Khim. Zh. (Russ. Ed.) 2007, 73 (7-8),

116-120.

189. Ismail, A. F.; Lau, W. J., Theoretical studies on structural and electrical properties of

PES/SPEEK blend nanofiltration membrane. AIChE J. 2009, 55 (8), 2081-2093.

86

190. Lau, W. J.; Ismail, A. F., Theoretical studies on the morphological and electrical properties of

blended PES/SPEEK nanofiltration membranes using different sulfonation degree of SPEEK. J. Membr.

Sci. 2009, 334 (1+2), 30-42.

191. Lau, W.-J.; Ismail, A. F., Effect of SPEEK content on the morphological and electrical properties

of PES/SPEEK blend nanofiltration membranes. Desalination 2009, 249 (3), 996-1005.

192. Yamada, S.; Kawabe, N. Aromatic polysulfone gas separation membranes. JP01242123A, 1989.

193. Rao, V.; Friedrich, K. A.; Stimming, U. In Proton-conducting membranes for fuel cells, CRC

Press: 2009; pp 759-820.

194. Paul, M.; Park, H. B.; Freeman, B. D.; Roy, A.; McGrath, J. E.; Riffle, J. S., Synthesis and

crosslinking of partially disulfonated poly(arylene ether sulfone) random copolymers as candidates for

chlorine resistant reverse osmosis membranes. Polymer 2008, 49 (9), 2243-2252.

195. Kim, Y.-J.; Lee, K.-S.; Jeong, M.-H.; Lee, J.-S., Highly chlorine-resistant end-group crosslinked

sulfonated-fluorinated poly(arylene ether) for reverse osmosis membrane. J. Membr. Sci. 2011, 378 (1-2),

512-519.

196. Kinzer, K. E.; Lloyd, D. R.; Gay, M. S.; Wightman, J. P.; Johnson, B. C.; McGrath, J. E., Phase-

inversion sulfonated polysulfone membranes. J. Membr. Sci. 1985, 22 (1), 1-29.

197. Blanco, J. F.; Nguyen, Q. T.; Schaetzel, P., Novel hydrophilic membrane materials: sulfonated polyethersulfone Cardo. Journal of Membrane Science 2001, 186 (2), 267-279.

198. Blanco, J.-F.; Sublet, J.; Nguyen, Q. T.; Schaetzel, P., Formation and morphology studies of

different polysulfone-based membranes made by wet phase inversion process. J. Membr. Sci. 2006, 283

(1+2), 27-37.

199. Rose, J. B. Sulfonated polyarylethersulphone copolymers. EP8894A1, 1980.

200. Rose, J. B. Sulfonated polyaryletherketones. EP8895A1, 1980.

201. Rose, J. B. Sulfonated poly(aryl ether ketones). EP41780A1, 1981.

202. Al-Omran, A.; Rose, J. B., Synthesis and sulfonation of poly(phenylene ether ether sulfone)s

containing methylated hydroquinone residues. Polymer 1996, 37 (9), 1735-1743.

203. Li, Y.; VanHouten, R. A.; Brink, A. E.; McGrath, J. E., Purity characterization of 3,3'-

disulfonated-4,4'-dichlorodiphenyl sulfone (SDCDPS) monomer by UV-vis spectroscopy. Polymer 2008, 49 (13-14), 3014-3019.

204. Geise, G. M.; Park, H. B.; Sagle, A. C.; Freeman, B. D.; McGrath, J. E., Water permeability and

water/salt selectivity tradeoff in polymers for desalination. J. Membr. Sci. 2011, 369 (1-2), 130-138.

205. Geise, G. M.; Paul, D. R.; Freeman, B. D., Fundamental water and salt transport properties of

polymeric materials. Prog. Polym. Sci. 2014, 39 (1), 1-42.

206. Xie, W.; Ju, H.; Geise, G. M.; Freeman, B. D.; Mardel, J. I.; Hill, A. J.; McGrath, J. E., Effect of

Free Volume on Water and Salt Transport Properties in Directly Copolymerized Disulfonated

Poly(arylene ether sulfone) Random Copolymers. Macromolecules (Washington, DC, U. S.) 2011, 44

(11), 4428-4438.

207. Lee, C. H.; McCloskey, B. D.; Cook, J.; Lane, O.; Xie, W.; Freeman, B. D.; Lee, Y. M.;

McGrath, J. E., Disulfonated poly(arylene ether sulfone) random copolymer thin film composite

membrane fabricated using a benign solvent for reverse osmosis applications. Journal of Membrane

Science 2012, 389 (0), 363-371.

208. Lee, C. H.; Spano, J.; McGrath, J. E.; Cook, J.; Freeman, B. D.; Wi, S., Solid-state NMR

molecular dynamics characterization of a highly chlorine-resistant disulfonated poly(arylene ether

sulfone) random copolymer blended with poly(ethylene glycol) oligomers for reverse osmosis

applications. J. Phys. Chem. B 2011, 115 (21), 6876-6884.

209. Xie, W.; Park, H.-B.; Cook, J.; Lee, C. H.; Byun, G.; Freeman, B. D.; McGrath, J. E., Advances

in membrane materials: desalination membranes based on directly copolymerized disulfonated

poly(arylene ether sulfone) random copolymers. Water Sci. Technol. 2010, 61 (3), 619-624.

210. Xie, R. J.; Gomez, M. J.; Xing, Y. J., Understanding permeability decay of pilot-scale

microfiltration in secondary effluent reclamation. Desalination 2008, 219 (1–3), 26-39.

87

211. McGrath, J. E. In Chlorine resistant membranes for reverse osmosis and nanofiltration,

American Chemical Society: 2009; pp POLY-224.

212. Sundell, B. J.; Lee, K.-s.; Cook, J. R.; Shaver, A.; Jang, E. S.; Freeman, B. D.; McGrath, J. E. In

Effect of backbone structure, degree of sulfonation and crosslinking on water transport properties for sulfonated poly(arylene ether sulfone) copolymers, American Chemical Society: 2014; pp POLY-487.

213. Hou, J.; Li, J.; Madsen, L. A., Anisotropy and Transport in Poly(arylene ether sulfone)

Hydrophilic-Hydrophobic Block Copolymers. Macromolecules 2010, 43 (1), 347-353.

214. Zhang, H.; Fang, S.; Ye, C.; Wang, M.; Cheng, H.; Wen, H.; Meng, X., Treatment of waste

filature oil/water emulsion by combined demulsification and reverse osmosis. Separation and

Purification Technology 2008, 63 (2), 264-268.

215. Ebewele, R. O., Polymer science and technology. CRC Press: 2000.

216. Sawyer, L. C.; Grubb, D. T., Polymer microscopy. Chapman and Hall: London; New York, 1996.

88

2.0 Disulfonated Poly(arylene ether sulfone) Random Copolymer Blends Tuned for Rapid

Water Permeation via Cation Complexation with Poly(ethylene glycol) Oligomers

Chang Hyun Leea, Ozma Lanea, James E. McGratha, Jianbo Houa, Louis A. Madsena, Justin

Spanoa, Sungsool Wia, Joseph Cookb, Wei Xieb, Geoffrey Geiseb, and Benny Freemanb*

a Macromolecules and Interfaces Institute, Virginia Polytechnic Institute And State University,

Blacksburg, VA 24061

bDepartment of Chemical Engineering, Center for Energy and Environmental Resources,

University of Texas at Austin, Austin, TX 78758

Reproduced with permission from Chemistry of Materials1

Here we present fundamental studies of a new blending strategy for enhancing water

permeability in ionomeric reverse osmosis membrane materials. A random disulfonated

poly(arylene ether sulfone) copolymer containing 20 mol percent hydrophilic units (BPS-

20) in the potassium salt form was blended with hydroxyl-terminated poly(ethylene glycol)

oligomers (PEG, Mn= 600-2000 g/mol) to increase the water permeability of BPS-20.

Blending PEG with the copolymer resulted in pseudoimmobilization of the BPS-20

polymers chains because PEG complexes with cations in the sulfonated polymer matrix.

Strong ion-dipole interactions between the potassium ions of the BPS-20 sulfonate groups

(-SO3K) and the PEG oxyethylene (-CH2CH2O-) were observed via NMR spectroscopy.

These interactions are similar to those reported between crown ethers and free alkali metal

systems. The PEG oligomers were compatible with the copolymer at 30 oC in an aqueous

environment. Transparent and ductile BPS-20/PEG blend films exhibited a Fox-Flory-like

89

glass transition temperature depression as the PEG volume fraction increased. This

depression depended on both PEG chain length and concentration within the BPS-20 matrix.

Both ion-dipole interactions and high coordination of –CH2CH2- with –SO3K yielded a

defined and interconnected hydrophilic channel structure. The water permeability and free

volume of BPS-20/PEG lend films containing 5 or 10 wt% PEG increased relative to BPS-

20. The blend films, however, exhibited reduced sodium choride (NaCl) rejection compared

to BPS-20. Addition of PEG did not significantly alter the material’s mechanical properties

in the dry or hydrated states. Unlike commercial state-of-the-art polyamide RO membranes,

the blend materials do not degrade when exposed to aqueous chlorine (hypochlorite) at pH

4. This comprehensive suite of measurements provides an understanding of the molecular

and morphological features needed for rational design of next-generation, chlorine-tolerant

water purification materials.

2.1 Introduction

The ability to efficiently and economically produce fresh water from brackish and seawater

is critical to addressing the growing global water shortage.2 Reverse osmosis (RO) processes can

be used to effective remove salts and large solutes, including bacteria, from natural water using

membranes with high flux and high salt rejection characteristics.3, 4 Currently, interfacially

polymerized aromatic polyamide (PA) thin film composite membranes are the state-of-the-art RO

technology because they offer high water flux and salt rejection (>99%) over a wide pH range.5

PA membranes, however, exhibit poor resistance to degradation by chlorine-based chemicals such

as sodium hypochlorite, which is used to disinfect water.6, 7 As a result, RO membrane performance

degrades when PA membranes are exposed to even low concentrations of available chlorine.6, 8, 9

90

Disulfonated poly(arylene ether sulfone) copolymers, now being developed for

desalination membrane applications, are more stable in chlorinated solution than commercial PA

membranes.10, 11 This result may be due to the absence of amide bonds in the sulfonated

polysulfone; amide bonds, such as those found in commercial PA RO membranes, are vulnerable

to attack by chlorine-based chemicals.12, 13 Figure 2-1 presents the chemical structure of 20 mol %

disulfonated poly(arylene ether sulfone) random copolymer (BPS-20). This copolymer, when

studied as a cast film, exhibits stable RO performance, i.e., no substantial decrease in salt rejection

or increase in water flux, after continuous exposure to chlorinated water at both high and low pH

(>40 h at 500 ppm chlorine).10 In contrast, a commercial PA membrane (SW30HR, Dow Film-

Tec) experienced a 20% decrease in salt rejection within 20 h of exposure to the same chlorinated

conditions.10 The water flux and salt rejection of BPS random copolymers exhibit a trade-off

relationship; highly water-permeable BPS materials tend to exhibit relatively low NaCl rejection

and vice versa. BPS-20 has been identified as an alternative candidate for desalination membrane

applications because its NaCl rejection is >99%.10 However, the BPS-20 water permeability

(0.033L m m-2h-1bar-1) is low. To put this value in perspective, consider a scenario where the

BPS-20 material has a 100 nm active layer in a composite membrane structure, similar to that of

commercially available state-of-the-art PA RO membranes. In this case, the water permeance

would be expected to be 0.33 L m-2 h-1 bar-1 (2000 ppm NaCl feed at 27.6 bar and pH 8). This

value can be compared to a DOW SW30HR RO membrane, whose performance is 1.1 L m-2 h-1

bar-1.14

This paper describes a new approach to increase the water permeability of BPS-20 and thus

improve the material’s desalination performance, via the addition of poly(ethylene glycol) (PEG,

Figure 2-1), a hydrophilic cation complexing agent. PEG oligomers were blended with BPS-20 to

91

tailor the resulting blend’s water permeability without changing the degree of sulfonation of the

BPS material. PEG systems have been widely used in gas separations,15, 16 water purification,17-21

and biomedical applications.22

Figure 2- 1. Pseudoimmobilization of PEG molecules with BPS-XX (Mþ = Naþ or Kþ, XX = degree of

sulfonation in mol %). In BPS-20, x is 0.2.

The water-soluble nature of PEG may limit its application in aqueous systems. PEG is often

immiscible with many polymers because it does not interact favorably with many polymer

matrices. As a result, water often extracts PEG from blends with many polymers. To prevent such

leaching, PEG chains can be immobilized via chemical modification or hydrogen bonding with

acidic polymers.18, 23 There are, however, applications for PEG where blending has been

successfully employed.15, 24

PEG ether oxygen atoms form complexes with a variety of metal cations (Li+, Na+, K+,

Cs+, and Rb+) via ion-dipole interactions similar to the behavior of cyclic ethers.25-27 Such

interactions in aqueous environments have been suggested to explain the miscibility of PEG and

salt form sulfonated polymers, which are composed of sulfonate anions and metal cations (-SO3M,

92

where M is a cation of an alkali metal element such as Na or K).28 If PEG does complex with metal

cations in sulfonated polymers, physical enthalpic interactions might effectively immobilize PEG

in the sulfonated polymer matrix. If PEG does complex with metal cations of sulfated polymers,

physical enthalpic interactions might effectively immobilize PEG in the sulfonated polymer matrix

without the use of covalent bonds (pseudoimmobilization, Figure 2-1). This interactions could

prevent PEG from leaching out of the polymer matrix upon exposure to water provided that the

interactions are strong and sustained. Additionally, PEG may increase the water permeability of

the sulfonated polymer matrix. This study seeks to understand the nature of the interaction between

PEG and the disulfonated copolymer, BPS-20. The main objective is to systematically investigate

the influence of PEG complexing agents, with different molecular weights and concentrations, on

the physicochemical characteristics of the salt form of sulfonated polymers membranes. To probe

these polymer blends, we employed pulsed-field gradient stimulated echo (PGSTE) NMR

spectroscopy, which can track diffusion of water molecules in mixed matrices over time using

magnetic field gradients to label nuclear spins with NMR frequencies based on their locations.29

Another major objective is to verify the efficacy of our pseudoimmobilization approach for

forming fast water transport pathways for desalination membranes. Finally, the chlorine resistance

of the blend films was compared to a PA membrane.

2.2 Experimental

2.2.1 Materials

BPS-20 in the potassium salt form was synthesized by Akron Polymer Systems (Akron,

OH) following published procedures.30-35 The material used in this study, BPS-20 (the exact

degree of sulfonation was 20.1 mol %, measured using 1H NMR) has an intrinsic viscosity of 0.82

dL g-1 in NMP with 0.05 M LiBr at 25oC. PEG oligomers (molecular weight Mn= 600 (0.6k), 1000

93

(1k), and 2000 (2k) with hydroxyl terminal endgroups (Figure 2-1) were purchased from Aldrich

Chemical Co. and were used as received. Dimethylacetamide (DMAc) (Aldrich Chemical Co.)

was used as a casting solvent without additional purification.

2.2.2 BPS-20/PEG Blends

After 2 g of BPS-20 was completely dissolved in DMAc at 30 oC, a PEG oligomer of the

desired molecular weight was added to the BPS-20 solutions in two different concentrations, 5 wt

% and 10 wt %, where wt % was defined relative to the mass of BPS-20 in the solution. The

resulting solutions were mixed for 1 day. Each solution (total solids concentration, 10 wt% in

DMAc) was degassed under vacuum at 25 oC for at least 1 day and cast on a clean glass plate.

Then the cast solution was dried for 4 h at 90 oC and heated to 150 oC for 1 day under vacuum.

The resulting films were easily peeled off of the glass plate and stored in deionized water at 30 oC

for 2 days to further remove residual solvent. The nominal thickness of all films was approximately

30-40 m, except those used for FT-IR measurement (20m). Transparent, ductile, and light

yellow BPS-20/PEG blend films were obtained. The yellow color increased with increasing PEG

molecular weight and concentration. The BPS-20/PEG films are denoted as BPS-20_PEG

molecular weight_PEG concentration (wt%). For example, BPS-20_PEG0.6k-5 denotes a BPS-20

film containing 5 wt % of 0.6 kDa PEG.

2.2.3 Characterization

The thermal decomposition of BPS-20_PEG films was investigated using a

thermogravimetric analyzer (TGA) (TA Instruments Q500 TGA) operated at a heating rate of 10

oC min-1 from 50 to 600 oC in a 60 mL min-1 nitrogen sweep gas. Prior to the thermal decomposition

measurements, all films were preheated in the TGA furnace at 110 oC for 15 min.

94

Transmission Fourier Transform Infrared (FT-IR) spectroscopy was used to study the

interactions between BPS-20 and PEG. FT-IR spectra in the range of 4000-900 cm-1 were obtained

using a Tensor 27 spectrometer (Bruker Optics).

Cross-polarization magic-angle spinning (CPMAS) 13C NMR spectra were taken with a

Bruker Avance II 300 MHz wide-bore spectrometer operating at Larmor frequencies of 75.47 MHz

for 13C and 300.13 MHz for 1H nuclei. Thin films samples (50-60 mg) were cut into small pieces

and packed into 4 mm magic angle spinning (MAS) rotors. Cross-polarization for 1 ms mixing

time was achieved at 50 kHz rf-field at the 13C channel with the 1H rf field ramped linearly over a

25% range centered at 38 kHz. A pulse technique known as total suppression of spinning side

bands (TOSS) was combined with a CP sequence to obtain sideband-free 13C MAS spectra at a 6

kHz spinning speed.36 The NMR signal averaging was achieved by coadding 2048 transients with

a 4 s acquisition delay time. 1H and 13C /2 pulse lengths were 4 and 5 s, respectively. Small

phase incremental alternation with 64 steps (SPINAL-64) decoupling sequence at 63 kHz power

was used for proton decoupling during 13C signal detection.37

To determine the glass transition temperature (Tg) of BPS-20 and BPS-20/PEG blend films,

dynamic mechanical analysis (DMA) was conducted using a TA DMA 2980 (TA Instruments) in

thin film tension mode; the temperature range was 0 to 300 oC with a ramp of 5 oC min-1 in a

nitrogen atmosphere. The films samples were 4 mm in width, and each was subjected to a preload

force of 0.025 N with an amplitude of 25 m at a frequency of 1 Hz.

Water uptake (%) was calculated using the following equation, where WD and WW are the

measured masses of dry and fully hydrated film samples, respectively. Each sample, approximately

95

5 cm x 5 cm, was dried in a vacuum oven at 110 oC for 1 day before measuring WD and immersed

in deionized water at 25 oC for 1 day before measuring WW.

𝑊𝑎𝑡𝑒𝑟 𝑈𝑝𝑡𝑎𝑘𝑒 = 𝑊𝑤 − 𝑊𝐷

𝑊𝐷∗ 100

Surface morphologies were examined via tapping mode atomic force microscopy (AFM),

using a Digital Instruments MultiMode scanning probe microscope with a NanoScope Iva

controller. A silicon probe (Veeo, end radius < 10 n, with a force constant k= 5 N m-1) was used

to image the samples, and the set point ratio was 0.82. Prior to measurement, all samples were

equilibrated to 30 oC and 40% relative humidity (RH) for at least 12 h.

For pulsed field gradient NMR spectroscopy, each film was cut into 5 x 5 mm peacies and

stacked together to a total mass of about 40 mg in a custom-built Teflon cell that was sealed to

maintain water content during diffusion measurements. The test cell was loaded into a Bruker

Avance III WB 400 MHz NMR spectrometer equipped with both a Micro5 triple-axis-gradient

(maximum 300 G cm-1) microimaging probe and an 8 mm double resonance (1H/2H) rf coil. The

pulsed-gradient stimulated echo pulse sequence (PGSTE) was applied with a /2 pulse time of 32

s, a gradient pulse duration () ranging from 1 to 3 ms, and diffusion time () ranging from 20

to 800 ms.29 Each measurement was repeated with 32 gradient steps, and the maximum gradient

strength was chosen to achieve 70-90% NMR signal attenuation.

The water permeability (Pw, L m m-2 h-1 bar -1) of BPS-20 and BPS-20_PEG films was

evaluated at 25 oC using a dead-end cell apparatus with a feed of 2000 ppm in deionized water. Pw

was defined as the volume of water (V) permeated per unit time (t) through a membrane sample

of area (A) and thickness (l) at a pressure difference (P = 400 psig):

96

𝑃𝑤 = 𝑉𝑙

𝐴𝑡∆𝑃

Salt rejection (R, %) was measured using a dead-end cell filtration apparatus with an

aqueous feed solution containing 2000 ppm NaCl at pH 6.5-7.5 and a pressure of 400 psig. Salt

rejection (R) was calculated as follows, wherein C f and Cp are the NaCl concentrations in the feed

and permeate, respectively. Salt concentration was measured with a NIST-traceable expanded

digital conductivity meter (Oakton Con 110 conductivity and TDS meter).

𝑅 = 𝐶𝑓−𝐶𝑝

𝐶𝑓∗ 100

Tensile properties of the films were determined using an Instron 5500R universal testing

machine equipped with a 200 lb load cell at 30 oC and 44-54% RH. Crosshead displacement speed

and gauge length were set to 5 mm min-1 and 25 mm, respectively. Dogbone specimens (50 mm

long and at least 4 mm wide) were cut from a single film. Prior to the measurement, each specimen

was dried under vacuum at 110 oC for at least 12 h, and then equilibrated at 30 oC and 4% RH.

2.3 Results and Discussion Because PEG is water-soluble and these materials are being evaluated for desalination

applications, the stability of the hydrated films was explored. To determine whether PEG leached

from the blend films, samples were stored in 30 oC deionized water for 150 days. After the 150

day soaking period, PEG content in the blend films was investigated using TGA, FT-IR, and NMR

spectroscopy. Figure 2-2 presents dynamic TGA thermograms of BPS-20 and BPS0-20/PEG blend

films. All of the materials exhibit three distinct thermal decomposition steps: (I) thermal

97

evaporation of water molecules [<215 oC], (III) thermal desulfonation of BPS-20 [375-420 oC],

and (IV) thermo-oxidation of BPS-20.35, 38 The initial weight loss was ascribed to desorption of

water from the samples; this weight loss increased with PEG concentration suggesting that water

hydrates both the PEG molecules and BPS-20 sulfonate groups.

An additional thermal decomposition step (II), which is ascribed to thermo-oxidation of

PEG [215-375 oC], was observed for the BPS20-Peg blend samples. Thermal decomposition of

PEG began around 215 oC, this temperature is higher than the initial thermal decomposition

temperature (Td) of pure PEG (~175 oC) and similar to the Td of the ester bridge grafted PEG.39, 40

This increase in the initial decomposition temperature suggests that PEG interacts with BPS-20.

Interactions of this nature may have bond energies similar to a weak covalent bond, such as an

ester.40 PEG decomposition was quantified and compared to the amount of PEG initially added to

the BPS-20 polymer matrix. The mass of PEG that remained in the blend matrix after the water

soaking step was essentially equal to the mass of PEG that was initially present in the blend.

Therefore, PEG did not leach from the blend matrix under our test conditions. We believe that the

Figure 2-2. TGA Thermograms of BPS-20 and BPS-20/PEG blends after soaking in deionized water for 150 days.

98

physical interaction between PEG and BPS-20 may arise from two sources: bonding interactions

between the BPS-20 sulfonate groups and PEG –OH groups and ion-dipole interactions between

PEG and the metal cation (K+) associated with the BPS-20 sulfonate groups. The maximum BPS-

20 thermal desulfonation temperature (TDS ~398 oC) decreases by 5-12 oC upon addition of PEG.

The reduction of TDS was more significant for the BPS-20/PEG blends containing higher molecular

weight PEG and higher PEG concentration.

Figure 2-3 presents FT-IR spectra of BPS-20 and BPS-20/PEG samples over the relevant

range of vibration frequencies to confirm the identity of the interactions between PEG and BPS-

20. To ensure that peak intensities were normalized, dry films of equivalent thicknesses (20 m)

were analyzed. The strong band at 2850 cm-1 in Figure 2-3 (a) is assigned to the stretching vibration

of the aliphatic alkyl PEG groups. The peak intensity of these groups increased with PEG

concentration, but the peak position did not shift. The bands at 1075 and 1030 cm-1 in Figure 2-3

(b) are attributed to the symmetric stretching vibration of sulfonate ions in BPS-20. The absorption

band at 1107 cm-1 is associated with the (-SO3K) asymmetric stretching vibration. Its frequency is

higher than the frequency of the corresponding vibration for the –SO3Na form of the same BPS-

20 polymer.41, 42 After the addition of the PEG, most of the peaks, including a stretching vibration

band of diphenyl ether at 1006 cm-1, did not shift. This result indicates that hydrogen bonding

between BPS-20 and PEG is not significant when the sample is in the dry state. In this discussion,

the relative intensity changes in the SO3- bands were excluded because of the presence of a strong

characteristic PEG band occurring between 1020 and 1200 cm-1.43

99

Figure 2-3. FT-IR spectra of BPS-20 and BPS-20_PEG materials with different concentrations (a and b) and

molecular weights (c) of PEG. (d) Simulated cation binding with PEG molecules (M+ = Na+, K+, and other cations).

In contrast, the absorption band (~950 cm-1) in Figure 2-3b, assigned to the PEG aliphatic

ether, became more distinct and shifted to higher frequency as PEG concentration increased. An

analogous band shift was observed in BPS-20/PEG samples containing high molecular weight

PEG (Figure 2-3c). The peak shift suggests that a chemical species exists in the vicinity of the

PEG molecules and physically interacts with the PEG aliphatic ether groups.

Free alkali metal cations, such as Na+ and K+, form complexes with PEG repeat units in

both aqueous and non-aqueous solvents.26, 44-46 Furthermore, the ion-dipole interaction of PEG

with the free metal cations is strengthened when long PEG chains (>9 repeat units) are used.47 In

particular, the selectivity of long PEG chains to potassium ions are promoted to the equivalent

100

level of crown ethers.28 PEG chains with more than 14 repeat units used in this study may interact

strongly with the potassium ions that are associated with the potassium sulfonate groups on the

BPS-20 chains (Figure 2-3d). This is because the ionic bond strength of the potassium sulfonate

group is theoretically stronger than the ion-dipole interaction between PEG and the sulfonate group

metal cations and the PEG repeat units have higher potassium coordination numbers (6-7)

compared to sodium ions (2-4).48, 49

Solid state 13C NMR spectroscopy of BPS-20/PEG blends provided information about the

interactions of BPS-20 and PEG (Figure 2-4). The solid state 13C NMR used here was sensitive

enough to monitor infinitesimal changes in the environment of the fully hydrated polymer system.

Characteristic peaks for hydrated BPS-20 were consistently observed at the same chemical shifts

regardless of PEG addition. Therefore, hydrogen bonding is not significant in hydrated BPS-20,

and the ion-dipole interaction governs the macroscopic properties of both the dry and hydrated

BPS-20/PEG system.

101

Figure 2-4. Solid state 13C NMR spectra of (a) BPS-20_PEG 0.6k-5 and (b) BPS-20.

We believe that the ability of PEG to complex with metal cations affected the BPS-20/PEG

blend glass transition temperature (Tg). Generally, the Tg of sulfonated polymers increases with

the degree of sulfonation due to the bulky and ionic nature of the sulfonate groups.50 For example,

the BPS-20 Tg (270 oC, Figure 2-5), is higher than that of BPS-00 (Radel, Tg = 220 oC). Also, BPS-

20 displays a broad Tg range, since the sulfonate groups are randomly distributed and may form

ionic domains of different sizes within the hydrophobic matrix. When incorporated into BPS-20,

PEG may disrupt the intermolecular and/or intramolecular ionic interactions between the sulfonate

groups in the BPS-20 ionic domains (entropic effect); a decrease in Tg could indicated this

102

disruption. A broad tan peak between 150 and 200 oC was observed using DMA. This broad

peak can be attributed to sulfonated ionic domain dilution that occurs when PEG is introduced to

BPS-20. Glass transition temperature depression behavior was especially significant in BPS-

20/PEG blend samples containing long PEG chains and higher PEG concentration. For example,

the Tg of BPS-20_PEG2k_10 dropped by 43 oC, comparable to that of nonsulfonated BPS-00. The

Tg of BPS-20/PEG blend samples decreased linearly with a slope dependent on PEG molecular

weight, which makes it possible to estimate the theoretical Tg change in BPS-20 upon PEG

addition.

BPS-20/PEG blend samples are binary systems composed of BPS-20 and PEG, which has

a Tg of -60 oC. The PEG homopolymer Tg is effectively contant over the molecular weights chosen

Figure 2-5. DMA profiles of BPS-20/PEG blends with different (a) molecular weights and (b) PEG concentrations.

103

for this study.51, 52 Unlike immiscible systems that show distinct and constant Tgs for each

component, the Tg of the BPS-20/PEG binary system depended on PEG concentration. The

measured glass transition temperatures are very similar to theoretical predictions made using the

Flory equation.53 This comparison is commonly used to determine whether binary systems are

miscible, compatible, or immiscible. In the following equation, Tg,i and W i represent the Tg and

weight fraction of component i, respectively.

1

𝑇𝑔,𝐵𝑃𝑆−20/𝑃𝐸𝐺=

𝑊𝐵𝑃𝑆−20

𝑇𝑔,𝐵𝑃𝑆−20+

𝑊𝑃𝐸𝐺

𝑇𝑔,𝑃𝐸𝐺

On the basis of the agreement between the measured BPS-20/PEG glass transition

temperature data and predictions made using the Flory-Fox equation, we conclude that BPS-20

and Peg form a compatible system as a result of ion-dipole interactions between PEG and the BPS-

20 sulfonate groups.

As the degree of sulfonation increases, the density of the BPS copolymer increases relative

to unsulfonated BPS-00 (1.30 g cm-3).54 When PEG is incorporated into BPS-20, the density of the

blended material decreases (Figure 2-6a). This decrease in density occurs because PEG acts as a

plasticizer that increases BPS-20 chain spacing and free volume. The BPS-20/PEG blend density,

when compared to the blend density calculated by assuming volume additivity, suggests that

blending PEG with BPS-20 results in free volume changes that extend beyond what would be

expected by simple mixing (Figure 2-6a). Dry BPS-20/PEG samples exhibit higher density than

104

wet samples. Density measurements also suggest that the free volume of BPS-20/PEG increased

as PEG chains became longer and more concentrated.

The water uptake of BPS-20/PEG correlated inversely with density (Figure 2-6b). Water

uptake increased as hydrophilic PEG was incorporated into BPS-20. Figure 2-6 indicates that low

density BPS-20/PEG samples (high free volume samples) showed greater water uptake than

samples of higher density (low free volume samples). Free volume is expected to influence water

and salt transport properties in these polymers.55

Figure 2-6. (a) Density and (b) water uptake of BPS-20/PEG blends. Calculated densities were obtained from volume

additivity.

105

In addition to water uptake, the surface morphology of sulfonated polymers can influence

water permeability. AFM images (Figure 2-7) indicate that the strong ion-dipole interaction of

PEG with potassium ions in BPS-20 sulfonate groups can induce a hydrophilic-hydrophobic

interaction even though BPS-20 is a random copolymer. In BPS-20, hydrophilic rod-like structures

indicated in the darker regions are randomly distributed throughout the light-colored hydrophobic

copolymer matrix.35 When PEG was added to BPS-20, the morphology of the matrix changed. In

BPS-20_PEG2k-5 (Figure 2-7b), hydrophilic phase connectivity and uniformity both appeared to

improve. This result may be related to strong ion-dipole interactions and the high coordination

number of PEG repeat units to potassium ions in the BPS-20 sulfonate groups.49, 56 PEG with more

than 9 repeat units typically prefers a meander or helical coil structure when exposed to free alkali

metal cations.47 The observed hydrophilic-hydrophobic phase separation depends on both PEG

chain length and concentration. As PEG chain length increased, at a constant concentration (10

wt%, Figure 2-7c-e), the hydrophilic domains became more interconnected but generally

decreased in size. Another morphological change was observed as the PEG concentration in BPS-

20/PEG2k was varied from 5 to 10 wt % (Figure 2-7b,e). Unlike BPS-20_PEG2k-5, BPS-

20_PEG2k-10 appears to have two different kinds of ionic domains: irregularly distributed ionic

domains, with sizes similar to BPS-20_PEG2k5, and capillary-shape ionic domains. The ionic

domains appear to be connect by long, tortuous hydrophilic pathways. The unevenly developed

morphology may cause BPS-20_PEG2k-10 to have a lower water permeability than BPS-

20_PEG2k-5.

106

Figure 2-7. AFM images of (a) BPS-20, (b) BPS-20_PEG2k-5, (c) BPS-20_PEG0.6k-10, (d) BPS-20_PEG1k-10, and (e)

BPS-20_PEG2k-10 materials in tapping mode. The dimensions of the images are 250 x 250 nm2. The phase scale is 0-20°. The measurement was conducted at 35% RH.

PGSTE-NMR provides information about the diffusion of molecules in materials, such as

the self-diffusion coefficient D of water in a polymer matrix. This technique is sensitive to the

identity of the mobile species and changes in the sample environment. D in a polymer matrix

depends strongly on water uptake and temperature. Morphological changes also strongly influence

the diffusion of water through the material. In fact, all samples exhibited reduced D values at long

diffusion times, indicating the existence of tortuous hydrophilic pathways similar to those observed

in the AFM images of Figure 2-7. However, interpreting the water permeation behavior in these

materials is not trivial since the PEG influences the hydrophilicity and ionic domain structure of

the material. In this study, we used the Mitra equation for porous media to assess diffusive

107

restrictions (related to the surface-to-volume ratio of diffusion pathways, S/V) that result from the

material’s morphology.57

𝐷 = 𝐷0(1 −4

9√𝜋 𝑆

𝑉√𝐷0∆)

Here, S/V (m-1) is a factor associated with the internal roughness in the mixed matrix. An

increase in S/V is expected to enhance water diffusion. By fitting D vs diffusion time , we

extracted values for the effective ‘free’ water diffusion coefficient D0 (that expected at very small

) and S/V. Here, D0 is interpreted as the effective intradomain diffusion coefficient of “free”

water through the polymer’s hydrophilic domain structure. Figure 2-8 shows the change in Do*S/V

as a function of PEG molecular weight. Do*S/V values appear to correlated with water

permeability data in Figure 2-9. This scaled water diffusion behavior in the mixed matrix materials

decreased with increasing PEG chain length. However, the Do*S/V values for the BPS-20/PEG

blend samples were greater than that for BPS-20. This finding indicates that S/V plays an important

role in the diffusion and permeation of water through these samples.

Figure 2-8. Diffusion behavior through tortuous water pathways in BPS-20/PEG 10% materials.

108

Figure 2-9 presents water permeability and salt rejection data for BPS-20/PEG films. In

the samples containing PEG, the water permeability of BPS-20/PEG increased relative to BPS-20.

The increase in water permeability depended on both PEG concentration and chain length. Water

permeability was greater in samples that contained a higher concentration of PEG; these samples,

with 10 wt % PEG, also exhibited higher water uptake. Water permeability of BPS-20/PEG,

however, decreased as PEG chain length increased. The water permeability of the BPS-20/PEG

blend films is likely affected by the morphological changes that occur upon adding PEG, as

indicated by AFM and scaled PGSTE-NMR (Mitra analysis). We believe that the morphological

contribution was particularly significant in BPS-20_PEG2k, which contained the highest PEG

molecular weight. The long and tortuous channels in BPS20_PEG2k-10 may restrict water

diffusion and cause the water permeability to decrease even though the water uptake increased

relative to that of BPS-20_PEG2k-5.

Figure 2-9. RO performance of BPS-20/PEG blend films: water permeability (left) and salt rejection (right).

The water permeability of the blend materials may also be influenced by hydrogen bonding

between water molecules and hydrophilic functional groups in the blend materials: BPS-20

sulfonate, PEG hydroxyl terminal endgroups, and PEG ether groups. A constant amount of BPS-

109

20 was used in BPS-20/PEG blends. Because the number of sulfonate groups was fixed, the

difference in hydrogen bonding activity within different BPS-20/PEG blend samples is

theoretically derived from the number of nonionic hydroxy and ether groups.58 Table 2-1 shows

the calculated number of the two functional groups per gram of BPS-20. The relative number of

hydroxyl groups increased as PEG molecular weight decreased. Furthermore, the relative number

hydroxyl groups increased when PEG concentration increased from 5 to 10 wt %. These results

are similar to the water permeability results in Figure 2-9, suggesting that the hydroxy groups may

contribute more to water permeability that the ether groups.

Table 2-1. Properties of BPS-20 and BPS-20/PEG blends

On the basis of the results shown in Figure 2-9, salt rejection decreases substantially with

increases in PEG concentration and molecular weight. For example, in a BPS-20 sample prepared

with 10 wt % of 2 kDa PEG, the rejection was 93.9%. For comparison, the rejection of pure BPS-

20 was 98.9%. We speculate that the ion-dipole interaction between K+ ions in the BPS-20

sulfonate groups and PEG oxyethylene units may weaken the electrostatic interaction of the

sulfonate groups that would typically form physically cross-linked, ion-selective domains.

Additionally, PEG incorportation into BPS-20 increases the material’s water uptake. This increase

in swelling reduces the concentration of sulfonate groups in the polymer matrix. This reduction in

the sulfonate group concentration may result in decreased ion exclusion and thus decrease salt

110

rejection.59 Both the disruption of ion-selective domains and the dilution of sulfonate group

concentration could result in reduced salt rejection as PEG concentration increases, which is

consistent with the experimental observations. The reduced salt rejection was more significant in

BPS-20/PEG samples prepared with high molecular weight PEG. Longer PEG chains are

associated with the formation of stronger ion-dipole interactions with K+ ions in the BPS-20

sulfonate groups and highly water swollen polymer matrices.

Plasticizers increase the free volume of a polymer matrix and weaken the inter- and

intramolecular interactions between the polymer chains; these effects often result in reduced

mechanical properties. In the samples containing PEG, the tensile modulus and strength were

unaffected by the blend composition (Table 2-1).

Figure 2-10. Solid state 13C NMR spectra of (a) PA and (b) BPS-20_PEG0.6k-5 after exposure to different

chlorine concentrations.

Resistance to degradation by chlorine-based disinfectants is critical for the long-term

performance of RO membranes. An accelerated chlorine stability test was conducted by immersing

111

each sample in a sealed vial containing a pH 4.0 +/- 0.3 buffered aqueous solution of sodium

hypochlorite (NaOCl) at concentrations of 100, 1000, and 10000 ppm.60 Each material was

immersed in the chlorinated solution for 2 time periods: 1 day and 1 week. Structural changes were

monitored with solid state 13C NMR spectroscopy (Figure 2-10). A reference PA film was obtained

from the interfacial polymerization of m-phenylenediamine (3 wt% in water) and trimesoyl

chloride (5 mM). The aromatic ring in the PA was vulnerable to electrophilic chlorine attack as

reported in the literature.61-63 The phenyl ring C-H peaks of the PA film decreased with exposure

to chlorine due to the formation of C-Cl bonds (Figure 2-10a). Also, the intensity of the carbonyl

carbon peak for the PA film decreased as PA chains were degraded by chlorine attack at the

carbonyl sites. In contrast, none of the BPS-20 peaks showed changes in position or intensity after

exposure at 10000 ppm chlorine for 1 week (Figure 2-10b). We note, however, that PEG

concentration in the blend materials decreases as chlorine concentration or exposure time

increases. For example, in BPS-20_PEG0.6k-5, the PEG content obtained from relative peak

integration with respect to the BPS20 peaks fell to about 60% of its initial value after 1 day of

immersion in 1000 ppm of chlorine. This may be related to oxidative degradation of PEG.64 No

additional PEG was lost when the films were exposed to highly concentrated solutions of chlorine

for a long time. A similar trend was observed for BPS20_PEG0.6k-10. This suggests that the ion-

dipole interaction between BPS-20 and PEG is stable under harsh conditions, although exposure

to chlorinated water may weaken the interaction.

2.4 Conclusions

Blends of PEG with BPS-20, a potassium salt form sulfonated random copolymer, gave

rise to strong ion-dipole interactions between potassium ions in the BPS-20 sulfonate groups and

PEG. These interactions are similar to the behavior seen in crown ethers and alkali metal cations.

These interactions resulted in high compatibility between BPS-20 and PEG and prevented PEG

112

from being extracted from the blends by exposure to water for long periods of time

(pseudoimmobilization). The strength of the ion-dipole interaction was similar to a weak covalent

bond, as shown in the thermal decomposition behavior of PEG in the blend materials. The cation

complexing capability of PEG molecules weakened the inter- or intramolecular hydrogen bonding

between sulfonate groups, which typically form physically cross-linked ionic domains. Increases

in PEG molecular weight and concentration resulted in a reduction of the blend’s Tg. This

plasticization led to increased free volume and increased water uptake. The ion-dipole interaction

and the high coordination number of the PEG repeat units to potassium ions in the BPS-20

sulfonate groups converted the sample surface morphology from a random distribution of

hydrophilic domains in the hydrophobic matrix into a more defined hydrophilic-hydrophobic

nanophase separated morphology. This trend was more pronounced in BPS-20/PEG blends

containing high concentrations of PEG. Increased water uptake and interconnected hydrophilic

domains, resulting from the addition of PEG, increased the water permeability of BPS-20/PEG

blends compared to BPS-20. Long PEG chains formed tortuous hydrophilic channels that

decreased water diffusion and thus water permeation. Furthermore, NaCl rejection decreased upon

the addition of PEG like because of weakened electrostatic interactions between potassium ions

and sulfonate groups, and reduced ionic exclusion due to dilution of the BPS-20 sulfonate groups

caused by increased water uptake. The decrease in NaCl rejection was minimized when short PEG

chains (e.g., 0.6k) were used. With the addition of PEG, water permeability increased to about

200% compared to the unblended BPS-20. The influence of PEG addition on sample toughness

and ductility was neglible. Unlike PA membranes, which degrade rapidly in the presence of

chlorine, BPS20/PEG blends resisted degradation after prolonged exposure to high chlorine

concentrations.

113

Incorporating PEG molecules into low disulfonated random copolymers offers and

effective and economical avenue to increase the material’s water permeability and fouling

resistance. However, the BPS-20 polymer matrix and hydroxyl-terminated PEG blends may not

exhibit the necessary water permeability and salt rejections to be an attractive RO membrane

material. Therefore, our ongoing studies are focusing on random copolymers with higher degrees

of sulfonation and ion-selective PEG. Finally, we will attempt to synthesize multiblock

copolymers containing PEG moieties to improve the hydrophilic and hydrophobic phase

separation of these chemically stable materials.

Acknowledgment. This work was supported by Dow Water and Process Solutions. This work was

also supported in part by the National Science Foundation (NSF)/Partnership for Innovation (PFI)

Program (Grant No. IIP-0917971). This material is based in part (L.A. Madsen and J. Hou) upon

work supported by the National Science Foundation under Award Number DMR 0844933.

Supporting information available: Glass transition temperature changes depending on PEG

molecular weight and concentration and stress-strain curves of BPS-20 and BPS-20/PEG materials

(PDF). This information is available free of charge via the internet at http://pubs.acs.org.

114

Reference:

1. Lee, C. H.; Van, H. D.; Lane, O.; McGrath, J. E.; Hou, J.; Madsen, L. A.; Spano, J.; Wi, S.; Cook,

J.; Xie, W.; Oh, H. J.; Geise, G. M.; Freeman, B. D., Disulfonated Poly(arylene ether sulfone) Random

Copolymer Blends Tuned for Rapid Water Permeation via Cation Complexation with Poly(ethylene

glycol) Oligomers. Chem. Mater. 2011, 23 (4), 1039-1049.

2. Service, R. F., Desalination Freshens Up. Science 2006, 313 (5790), 1088-1090.

3. Baker, R. W., Membrane Technology and Applications. 2004.

4. Geise, G. M.; Lee, H.-S.; Miller, D. J.; Freeman, B. D.; McGrath, J. E.; Paul, D. R., Water

purification by membranes: the role of polymer science. J. Polym. Sci., Part B: Polym. Phys. 2010, 48

(15), 1685-1718.



6. Knoell, T.; Martin, E.; Ishida, K.; Phipps, D. In The effects of chlorine exposure on the

performance and properties of polyamide reverse osmosis membranes, American Water Works

Association: 2005; pp 504-516.

7. Petersen, R. J.; Codotte, J. E.; Porter, M. C., Handbook of Industrial Membrane Technology.

1990; p 307.

8. Kucera, J., Reverse osmosis : design, processes, and applications for engineers. Scrivener Pub. ;

Wiley: Salem, Mass.; Hoboken, N.J., 2010.

9. Do, V. T.; Tang, C. Y.; Reinhard, M.; Leckie, J. O., Effects of Chlorine Exposure Conditions on

Physiochemical Properties and Performance of a Polyamide Membrane - Mechanisms and Implications. Environ. Sci. Technol. 2012, 46 (24), 13184-13192.

10. Park, H. B.; Freeman, B. D.; Zhang, Z.-B.; Sankir, M.; McGrath, J. E., Highly chlorine-tolerant

polymers for desalination. Angew. Chem., Int. Ed. 2008, 47 (32), 6019-6024.




12. Brousse, C.; Chapurlat, R.; Quentin, J. P., New membranes for reverse osmosis. I. Characteristics

of the base polymer: sulfonated polysulfones. Desalination 1976, 18 (2), 137-53.

13. Drzewinski, M.; Macknight, W. J., Structure and properties of sulfonated polysulfone ionomers.

J. Appl. Polym. Sci. 1985, 30 (12), 4753-70.

14. Van Wagner, E. M.; Sagle, A. C.; Sharma, M. M.; Freeman, B. D., Effect of crossflow testing

conditions, including feed pH and continuous feed filtration, on commercial reverse osmosis membrane

performance. J. Membr. Sci. 2009, 345 (1-2), 97-109.

15. Chakrabarty, B.; Ghoshal, A. K.; Purkait, M. K., SEM analysis and gas permeability test to

characterize polysulfone membrane prepared with polyethylene glycol as additive. J. Colloid Interface

Sci. 2008, 320 (1), 245-253.

16. Chakrabarty, B.; Ghoshal, A. K.; Purkait, M. K., Effect of molecular weight of PEG on

membrane morphology and transport properties. J. Membr. Sci. 2008, 309 (1+2), 209-221.

17. Ju, H.; McCloskey, B. D.; Sagle, A. C.; Kusuma, V. A.; Freeman, B. D., Preparation and

characterization of crosslinked poly(ethylene glycol) diacrylate hydrogels as fouling-resistant membrane

coating materials. J. Membr. Sci. 2009, 330 (1+2), 180-188.

18. Ju, H.; McCloskey, B. D.; Sagle, A. C.; Wu, Y.-H.; Kusuma, V. A.; Freeman, B. D., Crosslinked

poly(ethylene oxide) fouling resistant coating materials for oil/water separation. J. Membr. Sci. 2008, 307

(2), 260-267.

19. La, Y.-H.; McCloskey, B. D.; Sooriyakumaran, R.; Vora, A.; Freeman, B.; Nassar, M.; Hedrick,

J.; Nelson, A.; Allen, R., Bifunctional hydrogel coatings for water purification membranes: Improved

fouling resistance and antimicrobial activity. J. Membr. Sci. 2011, 372 (1-2), 285-291.

115

20. McCloskey, B. D.; Ju, H.; Freeman, B. D., Composite Membranes Based on a Selective

Chitosan-Poly(ethylene glycol) Hybrid Layer: Synthesis, Characterization, and Performance in Oil-Water

Purification. Ind. Eng. Chem. Res. 2010, 49 (1), 366-373.

21. Sagle, A. C.; Van Wagner, E. M.; Ju, H.; McCloskey, B. D.; Freeman, B. D.; Sharma, M. M.,

PEG-coated reverse osmosis membranes: Desalination properties and fouling resistance. J. Membr. Sci.

2009, 340 (1-2), 92-108.

22. Wang, P.; Tan, K. L.; Kang, E. T.; Neoh, K. G., J. Mater. Chem. 2001, 11, 783.

23. Revanur, R.; McCloskey, B.; Breitenkamp, K.; Freeman, B. D.; Emrick, T., Reactive amphiphilic

graft copolymer coatings applied to polyvinylidene fluoride ultrafiltration membranes. Macromolecules

(Washington, DC, U. S.) 2007, 40 (10), 3624-3630.

24. Chou, W. L.; Yu, D. G.; Yang, M. C.; Jou, C. H., Sep. Purif. Technol. 2007, 57, 209.

25. Gokel, G. W., Crown Ethers and Cryptands. 1991.

26. Okada, T., Macromolecules 1990, 23, 4216.

27. Doan, K. E.; Heyen, B. J.; Ratner, M. A.; Shriver, D. F., Chem. Mater. 1990, 2, 539.

28. Yanagida, S.; Takahasi, K.; Okahara, M., Bull. Chem. Soc. Jpn. 1977, 50, 1386.

29. Madsen, L. A.; Li, J.; Hou, J., Probing transport in ionomer membranes via NMR anisotropy and

diffusion measurements. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 2009, 50 (2), 788-789. 30. Li, Y.; VanHouten, R. A.; Brink, A. E.; McGrath, J. E., Purity characterization of 3,3'-

disulfonated-4,4'-dichlorodiphenyl sulfone (SDCDPS) monomer by UV-vis spectroscopy. Polymer 2008,

49 (13-14), 3014-3019.

31. Li, Y.; Wang, F.; Yang, J.; Liu, D.; Roy, A.; Case, S.; Lesko, J.; McGrath, J. E., Synthesis and

characterization of controlled molecular weight disulfonated poly(arylene ether sulfone) copolymers and

their applications to proton exchange membranes. Polymer 2006, 47 (11), 4210-4217.

32. Sankir, M.; Bhanu, V. A.; Harrison, W. L.; Ghassemi, H.; Wiles, K. B.; Glass, T. E.; Brink, A. E.;

Brink, M. H.; McGrath, J. E., Synthesis and characterization of 3,3'-disulfonated-4,4'-dichlorodiphenyl

sulfone (SDCDPS) monomer for proton exchange membranes (PEM) in fuel cell applications. J. Appl.

Polym. Sci. 2006, 100 (6), 4595-4602.

33. Kim, Y. S.; Dong, L.; Hickner, M. A.; Glass, T. E.; Webb, V.; McGrath, J. E., State of Water in Disulfonated Poly(arylene ether sulfone) Copolymers and a Perfluorosulfonic Acid Copolymer (Nafion)

and Its Effect on Physical and Electrochemical Properties. Macromolecules 2003, 36 (17), 6281-6285.

34. Sumner, M. J.; Harrison, W. L.; Weyers, R. M.; Kim, Y. S.; McGrath, J. E.; Riffle, J. S.; Brink,

A.; Brink, M. H., Novel proton conducting sulfonated poly(arylene ether) copolymers containing

aromatic nitriles. J. Membr. Sci. 2004, 239 (2), 199-211.

35. Wang, F.; Hickner, M.; Kim, Y. S.; Zawodzinski, T. A.; McGrath, J. E., Direct polymerization of

sulfonated poly(arylene ether sulfone) random (statistical) copolymers: candidates for new proton

exchange membranes. J. Membr. Sci. 2002, 197 (1-2), 231-242.

36. Dixon, W. T.; Schaefer, J.; Sefcik, M. D.; Stejskal, E. O.; McKay, R. A., Total suppression of

sidebands in CPMAS carbon-13 NMR. J. Magn. Reson. 1982, 49 (2), 341-5.

37. Fung, B. M.; Khitrin, A. K.; Ermolaev, K., An Improved Broadband Decoupling Sequence for

Liquid Crystals and Solids. J. Magn. Reson. 2000, 142 (1), 97-101.

38. Lee, C. H.; Park, H. B.; Chung, Y. S.; Lee, Y. M.; Freeman, B. D., Water Sorption, Proton

Conduction, and Methanol Permeation Properties of Sulfonated Polyimide Membranes Cross-Linked with

N,N-Bis(2-hydroxyethyl)-2-aminoethanesulfonic Acid (BES). Macromolecules 2006, 39 (2), 755-764.

39. Hechavarria, L.; Mendoza, N.; Altuzar, P.; Hu, H., In situ formation of polyethylene glycol-

titanium complexes as solvent-free electrolytes for electrochromic device application. J. Solid State Electrochem. 2010, 14 (2), 323-330.

40. Cappelli, A.; Galeazzi, S.; Giuliani, G.; Anzini, M.; Grassi, M.; Lapasin, R.; Grassi, G.; Farra, R.;

Depas, B.; Aggravi, M.; Donati, A.; Zetta, L.; Boccia, A. C.; Bertini, F.; Samperi, F.; Vomero, S.,

Macromolecules 2009, 42, 2368.

41. Zundel, G., Hydration and Intermolecular Interaction. 1969.

116

42. Zundel, G., Hydration structure and intermolecular interaction in polyelectrolytes. Angew. Chem., Int. Ed. Engl. 1969, 8 (7), 499-509.

43. Kim, C. H.; Kim, D. W.; Cho, K. Y., Polym. Bull. 2009, 63, 91.

44. Izatt, R. M.; Bradshaw, J. S.; Nielsen, S. A.; Lamb, J. D.; Christensen, J. J.; Sen, D.,

Thermodynamic and kinetic data for cation-macrocycle interaction. Chem. Rev. 1985, 85 (4), 271-339.

45. Izatt, R. M.; Bradshaw, J. S.; Pawlak, K.; Bruening, R. L.; Tarbet, B. J., Thermodynamic and

kinetic data for macrocycle interaction with neutral molecules. Chem. Rev. 1992, 92 (6), 1261-354.

46. Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L., Chem. Rev. 1991, 91, 1721.

47. Chan, K. W. S.; Cook, K. D., Macromolecules 1983, 16, 1736.

48. Kerres, J. A., Fuel Cells 2005, 5, 230.

49. Besner, S.; Vallee, A.; Bouchard, G.; Prud’homme, J., Macromolecules 1992, 25, 6480.

50. Xu, K.; Li, K.; Khanchaitit, P.; Wang, Q., Chem. Mater. 2007, 19, 5937.

51. Back, D. M.; Schmitt, R. L., Kirk-Othmer Encyclopedia of Chemical Technology. 2004; Vol. 10,

p 673.

52. Read, B. E., Polymer 1962, 3, 529.

53. Fox, T. G., Bull. Am. Phys. Soc. 1956, 1, 123.

54. Polyers, S. A. 55. Yasuda, H.; Lamaze, C. E.; Ikenberry, L. D., Die Makromol. Chem. 1968, 118, 19.

56. Philippova, O. E.; Karibyants, N. S.; Starodubtzev, S. G., Macromolecules 1994, 27, 2398.

57. Mitra, P. P., Physica A 1997, 241, 122.

58. Porter, M. R., Handbook of Surfactants. 1994.

59. Helfferich, F., Ion Exchange. 1995.

60. Jayarani, M. M.; Rajmohanan, P. R.; Kulkarni, S. S.; Kharul, U. K., Desalination 2000, 130, 1.

61. Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D., A comprehensive review on PEM water

electrolysis. Int. J. Hydrogen Energy 2013, 38 (12), 4901-4934.

62. March, J., Advanced Organic Chemistry: Reactions, Mechanisms, and Structure. 1992.

63. Bieron, J. F.; Dinan, F. J.; Zabicky, J., The Chemistry of Amides. 1970; p 263.

64. Chen, S. F.; Zheng, J.; Li, L. Y.; Jiang, S. Y., J. Am. Chem. Soc. 2005, 127, 14473.

117

3.0 Kinetics of Post-Sulfonation of a Poly(arylene ether sulfone) that

Contains Hydroquione (Radel-A) for Application in Reverse Osmosis

Membranes

Ozma Lane, S. J. Mecham, Eui Soung Jang, B. D. Freeman, J. S. Riffle, J. E. McGrath

Macromolecules and Interfaces Institute, Virginia Tech, Blacksburg, VA 24061

Abstract

Partially hydrophilic engineering thermoplastics have been investigated as potential

alternative materials for reverse osmosis. Post-sulfonated poly(arylene ether sulfone)s containing

hydroquinone are of particular interest as they feature facile synthesis with precise tailoring of

composition. This research focuses on optimizing post-sulfonation conditions for achieving

controlled sulfonation on the hydroquinone unit, while avoiding significant secondary reactions or

chain scission. It was found that the copolymer with 29% of the repeat units containing

hydroquinone could be efficiently sulfonated within an hour at 50 oC without significant chain

degradation. Sulfonation of the polymer was determined to be a function of the rate of powder

dissolution.

3.1 Introduction

Projections for increasing demand for potable water coupled with greater levels of

contamination in groundwater supplies require the development of increased capacity for potable

118

water generation.1 Reverse osmosis (RO) is more cost-effective than thermal methods of

desalination, and its market growth has outpaced that of its competitors.2 The current predominant

material used in RO production facilities is a crosslinked aromatic polyamide.3 While offering

excellent salt rejection and water flux, this material has two significant drawbacks. The interfacial

synthesis of the polyamide produces a rough surface, which affords both a greater surface area for

the adhesion of biofouling microbes as well as areas that are shielded from the shear force of the

feedstream. Together, these provide an environment conducive to biofouling.4-6 Secondly, the

amide bond of the crosslinked aromatic polyamide dense membrane is susceptible to attack by

chlorinated disinfectants, resulting in chain scission.7 To avoid membrane degradation, a

pretreatment process is required wherein the feedstream is first chlorinated to reduce microbial

levels, then dechlorinated with sodium bisulfite before being passed through the membrane to

avoid membrane degradation, and re-chlorinated after membrane desalination. A membrane that

is stable in the presence of these chlorinated disinfectants would reduce the time and expense

necessary for the separation, and it potentially could also reduce biofouling due to the introduction

of disinfectants at the membrane surface.8-11

Research is underway to develop an alternative membrane which lacks chlorine

susceptibility, while offering a smoother membrane surface and comparable water transport

properties. A number of studies have been performed on sulfonated poly(arylene ether sulfone)s,

demonstrating good chlorine resistance and smooth surfaces, with good transport properties.11-13

These materials have been synthesized from a combination of nonsulfonated and disulfonated

monomers. Some investigations have indicated that the partially disulfonated polymers feature

compromised sodium chloride rejection when calcium salts are present in the feedstream. It has

119

been proposed that the compromising effect of calcium is due to the small distance between the

sulfonic acid groups on the disulfonated poly(arylene ether sulfone)s.

A potential alternative is to prepare a monosulfonated poly(arylene ether sulfone) where

the sulfonation is conducted post-polymerization. Post-polymerization as an approach for

sulfonating poly(arylene ether)s was evaluated in the 1980s, and then abandoned due to poor

control over the extent of sulfonation, inability to control the microstructure of the sulfonated units,

decrease in molecular weight due to chain scission, and unstable sulfonic acid groups.14 While

most of the current synthetic investigations use the direct copolymerization of a disulfonated

monomer to produce novel hydrophilic materials, select cases in which post-sulfonation is a viable

and potentially advantageous method may still exist.

This work describes the copolymerization and post-sulfonation of polymers derived from

dichlorodiphenylsulfone with a systematic series of mixtures of hydroquinone and bisphenol S.

Post-sulfonation takes place almost exclusively on the hydroquinone rings since all of the other

rings are substituted with electron-withdrawing sulfone groups that are deactivated toward

electrophilic aromatic sulfonation. This avoids the issues associated with the microstructure of the

sulfonate ions since these copolymers should have the hydroquinone distributed randomly along

the chains. A series of investigations on this class of materials have been published confirming the

selectivity of this reaction, but consistent, detailed descriptions of the reaction kinetics and rates

of molecular weight degradation were not provided.15-17 In this current research, the reaction

kinetics and measurements of molecular weight degradation were studied to optimize the

sulfonation process with a minimal level of chain scission. This information is to be used as a

model study for developing a series of post-sulfonated polymers with varying structures in order

to determine their resistance to the compromising effects of calcium on sodium rejection.

120

3.2 Experimental

3.2.1 Materials

Radel-A was kindly provided by Solvay and used as received. The structure of Radel-A

is illustrated in Figure 3-1 below. It is a poly(arylene ether sulfone) with approximately 29% of

the repeat units containing hydroquinone (rather than bisphenol S). The precise composition of

hydroquinone was obtained from calculations using integrations of 1H NMR spectra. Concentrated

sulfuric acid (H2SO4) was obtained from VWR and used as received. N,N-dimethylacetamide was

used as received from Aldrich.

Figure 3-1. Structure of Radel-A.

3.2.2 Sulfonation of Radel A

Prior to the sulfonation reaction, a solution of 30% (w/v) Radel-A in dimethylacetamide

was made and precipitated in deionized water in a blender to provide a high surface area powder.

This facilitated rapid dissolution during the sulfonation reaction. The precipitated polymer was

filtered, washed with deionized water, dried without vacuum at 100 oC for 12 h and then under

vacuum at 110 oC for 12 h to remove the solvent.

For the sulfonation reaction, a four-necked flask equipped with an overhead stirrer,

nitrogen inlet, condenser and a thermometer adapter was assembled. An oil bath with a

thermocouple was used to control the reaction temperature. Radel A powder (15 g) and sulfuric

acid (150 mL) were added into the flask. Reactions were performed at 40, 50, and 60 oC. Time

zero was designated when the temperature reached the desired point for the kinetics experiment

121

(~2 min). Aliquots of 5-10 mL were removed at 5, 10, 15, 30, 60, and 120 min. The aliquots were

quenched by precipitation in deionized water, followed by washing with copious amounts of

deionized water until the pH reached at least 5.

3.2.3. 1H NMR

Samples were dried overnight at 110 oC in a vacuum oven. In a scintillation vial with

molecular sieves, approximately 10 mg of the polymer was dissolved in 700 µL of d6-DMSO.

Spectra were collected on a Varian Unity Plus spectrometer operating at 400 MHz. COSY

experiments were performed with 16 increments per second and 200 scans.

3.2.4. Water Uptake

Water uptake was measured gravimetrically. Polymer films (100 mg) were cast from a 10%

(w/v) solution in DMAc onto a clean glass plate and placed under an IR lamp for 6-12 h. Films

were removed via immersion in deionized water, and heated in water at 80 oC for 3 h to remove

any residual solvent. The samples were dried overnight at 110 °C under vacuum. Films were then

equilibrated in deionized water overnight, converted from the acid to their salt form by heating in

1.0 M NaCl at 80 oC for two h, then they were kept at room temperature in the water overnight.

They were blotted and weighed to obtain the wet weight (Wwet). Films were dried under vacuum

at 110 oC for 12 h to obtain dry weights (Wdry). The water uptake in the sodium salt form was

determined as follows:

𝑊𝑎𝑡𝑒𝑟 𝑈𝑝𝑡𝑎𝑘𝑒 =𝑊𝑤𝑒𝑡 − 𝑊𝑑𝑟𝑦

𝑊𝑑𝑟𝑦∗ 100%

3.2.5 Size Exclusion Chromatography

122

Size exclusion chromatography (SEC) was conducted on the polymers to measure

molecular weights and molecular weight distributions. The solvent was DMAc that was distilled

from CaH2 and that contained dry LiCl (0.10 M). The column set consisted of 3 Agilent PLgel 10-

μm Mixed B-LS columns 300x7.5 mm (polystyrene/divinylbenzene) connected in series with a

guard column having the same stationary phase. The column set was maintained at 50 °C. An

isocratic pump (Agilent 1260 infinity, Agilent Technologies) with an online degasser (Agilent

1260), autosampler and column oven was used for mobile phase delivery and sample injection. A

system of multiple detectors connected in series was used for the analyses. A multi-angle laser

light scattering (MALLS) detector (DAWN-HELEOS II, Wyatt Technology Corp.), operating at

a wavelength of 658 nm, a viscometer detector (Viscostar, Wyatt Technology Corp.), and a

refractive index detector operating at a wavelength of 658 nm (Optilab T-rEX, Wyatt Technology

Corp.) provided online results. The system was corrected for interdetector delay and band

broadening. Data acquisition and analysis were conducted using Astra 6 software from Wyatt

Technology Corp. Validation of the system was performed by monitoring the molar mass of a

known molecular weight polystyrene sample by light scattering. The accepted variance of the

21,000 g/mole polystyrene standard was defined as 2 standard deviations (11.5% for Mn and 9%

for Mw) derived from a set of 34 runs.

3.3 Results and Discussion

3.3.1. Post-sulfonation of Radel A

Radel-A is a poly(arylene ether sulfone) with approximately 29% hydroquinone-

containing comonomer, as shown in Figure 3-1. It is chemically resistant with good mechanical

properties, and was selected as a material for post-sulfonation kinetics in order to develop a

123

baseline for sulfonation parameters to be tested on materials with varying hydroquinone

compositions in later studies.

It was expected that sulfonation would predominantly occur on the hydroquinone rings of

the poly(arylene ether sulfone) because all of the other rings were deactivated toward electrophilic

aromatic substitution by the presence of the electron withdrawing sulfone linkages. The

compositions of the non-sulfonated starting material and the post-sulfonated samples were

investigated with 1H NMR. Figures 3-2 and 3-3 compare the chemical structures of the non-

sulfonated and post-sulfonated Radel-A after 2 hours of sulfonation at 60 oC. In the non-sulfonated

Radel-A, the A and A1 peaks resonate at 7.95 and 7.93 ppm and overlap somewhat. The B and B1

peaks resonate at 7.23 and 7.11 ppm and are better separated so that the relative number of units

adjacent to the hydroquinone can be distinguished. The C peak that corresponds to the

hydroquinone protons resonates as a singlet at 7.17 ppm. The amount of hydroquinone-containing

repeat units was calculated to be 29% (Figure 3-2) based on the integrals of the B1 and C peaks.

Protons on the post-sulfonated Radel-A structure in the 1H NMR spectrum are

distinguished from their non-sulfonated predecessors by the presence of a prime after the peak

labels (Figure 3-3). In the post-sulfonated Radel spectrum, the A’ and A1’ peaks resonate at 7.96

and 7.83 ppm, with the A1’ peaks somewhat shifted due to the proximity of the sulfonated

hydroquinone. The B’ and B1’ peaks resonated at 7.24 and 6.97 ppm, with the shift of the B1’

peak again reflecting the adjacent sulfonated hydroquinone rings. The small peak at 7.18 ppm is

attributed to residual non-sulfonated hydroquinone. The ratio of the B1’ to B’ integrals could not

be precisely determined during the course of the kinetics study due to overlap of the unreacted B,

B’, and D’ peaks. The sulfonated hydroquinone peaks C’, D’, and E’ resonate at 7.4, 7.15, and

124

7.05 ppm. The peak corresponding to the D’ proton also overlaps with neighboring peaks and

unreacted B and B1 proton peaks.

Figure 3-2. Structure and 1H NMR of the non-sulfonated Radel-A.

125

Figure 3-3. Structure of 1H NMR of sulfonated Radel-A with peak assignments.

In order to confirm the sulfonated copolymer structure, 2-D homonuclear correlational

spectroscopy experiments were performed to investigate correlations between neighboring

protons. In the COSY NMR (1H-1H) spectrum (Figure 3-4), the C’ proton (7.45 ppm) has a weak

interaction with the D’ proton, as indicated by the light blue lines identifying those protons and

their correlating peak on the COSY spectrum. This weak interaction is due to coupling across 4

bonds of the aromatic ring and protons, rather than the 3 bonds that are otherwise observed in this

126

spectrum. The D’ and E’ protons correlate strongly with each other, as highlighted by the black

lines in Figure 3-3. A’ and B’, as well as A1’ and B1’ protons, show a stronger correlation due to

a smaller number of bonds of separation, but there are no correlations or peaks suggesting a

secondary sulfonation site anywhere other than the hydroquinone repeat unit.

127

Figure 3-4. The COSY (1H-1H) spectrum of Radel-A after 2 hours of sulfonation at 60 oC. Black lines indicate

the proximity of the C and D protons, with no other correlations for the C proton.

3.3.2 Kinetics

After confirming the structure of the sulfonated Radel-A and the absence of significant side

reactions, the level of sulfonation at varying sulfonation temperatures and times was determined.

128

The spectrum of the polymer that was sulfonated for 2 hours at 60 °C was used to confirm the

degree of sulfonation of a fully sulfonated Radel-A polysulfone (Figure 3-3). The reaction was

considered complete since the peak corresponding to residual unsulfonated hydroquinone was

negligible. The ratio of the A’ and A1’ integrals to the C’ integral was used to determine the ratio

of hydroquinone-containing repeat units to bisphenol sulfone repeat units. The integrations were

normalized such that the C’ integral was set to 1. The integral of the A’ + A1’ peaks (21.8) was

divided by 2 to determine the number of rings proximate to sulfone groups (10.9). Two of the

sulfone rings were assigned to the hydroquinone-containing repeat unit, while the remainder of 8.9

were assigned to the bisphenol sulfone repeat unit. The number of rings next to a sulfone on the

hydroquinone-containing repeat unit was divided by two rings per repeat unit, giving one repeat

unit. The number of rings on the bisphenol sulfone repeat unit (8.9) were divided by 4 rings per

repeat unit, giving a relative number of 2.25 repeat units per each hydroquinone-containing repeat

unit. The percentage of hydroquinone-containing repeat units was determined as the number of

hydroquinone units divided by the sum of the hydroquinone-containing repeat units and bisphenol

sulfone repeat units (1/(1+2.25)). The percentage of bisphenol sulfone-containing repeat units was

determined as the number of bisphenol sulfone units divided by the sum of the hydroquinone-

containing repeat units and bisphenol sulfone repeat units (2.25/(1+2.25). The ratio was calculated

to be 30.8% sulfonated hydroquinone-containing repeat units, which is within the margin of error

of the 29% non-sulfonated hydroquinone determined from the 1H NMR spectrum of the original

Radel-A.

For partially sulfonated polymers collected during the course of sulfonation, the following

is a representative calculation for the percentage of sulfonation of the hydroquinone-containing

units. In the sample removed at 60 oC after 15 min of sulfonation, the integral of the A’+ A1’ peaks

129

was 36.85 and the integral of the C’ peak was normalized to 1. The A’+A1’ peak integral was

divided by 2 to provide the number of sulfone-proximate rings (18.4). The ratio of sulfone-

proximate rings in the hydroquinone-containing units relative to the bisphenol sulfone-containing

units is 0.43. From the number of calculated rings, 14.75 rings and 3.69 repeat units were attributed

the bisphenol sulfone-containing units and 3.25 rings and 1.63 repeat units were attributed to the

hydroquinone-containing comonomers. Therefore, 1/1.63 repeat units provided a sulfonation level

of 61%.

In Figure 3-5, the percentage of hydroquinone that was sulfonated as determined by 1H

NMR is shown as a function of sulfonation time and reaction temperature. It indicates that the

reaction is effectively complete after 2 hours at 50 oC. The rate of sulfonation is likely very

dependent on the rate of dissolution of the polymer powder. The reaction appears to change from

a cloudy mixture of suspended particles upon initial mixing, and becomes a clear dark amber by

the 120 minute mark.

130

Figure 3-3. Sulfonation of hydroquinone (%) as a function of reaction time and temperature.

3.3.2. Molecular weights

SEC results indicate that the sulfonation process does not produce significant degradation

within the sulfonation conditions studied (Table 3-1). It should be noted that according to NMR

calculations, the polymer collected after 120 minutes of sulfonation at 40 oC is not yet fully

sulfonated and thus, it was not evaluated further. The molecular weight results provide good

evidence that the reaction conditions utilized herein can avoid the molecular weight degradation

that has previously caused concern in other post-sulfonated polysulfones, where the sulfonation

took place on the biphenol group. The more rapid reaction of the hydroquinone allows for

sulfonation to proceed before chain degradation can take place. This provides a useful reference

for reaction conditions for the post-sulfonation of a polymer series with a variable hydroquinone

content, which may be of use in membrane separations applications.

131

Table 3-1. Mw of Radel A (g/mole) before and after post-sulfonation at 50 and 60 oC. Mw obtained by SEC in DMAc with

0.10 M LiCl.

Time (min) 50 oC 60 oC

0 35,000 35,000

60 42,600 36,800

120 56,600 32,300

3.4 Conclusions

The hydroquinone unit of the Radel-A™ polymer investigated in this study was rapidly

post-sulfonated in a sulfuric acid solution. This is a rapid, facile and quantitative method for mild

sulfonation of only the activated rings. Using these mild conditions, SEC results showed that the

sulfonation procedure did not degrade the polymers. Due to the structure of the polymer, this

ensures that the sulfonate ions cannot be on adjacent rings. Importantly, this spaces the ions

sufficiently far apart that chelation or binding to multivalent cations will not be likely due to

adjacent sulfonates.

1. Elimelech, M.; Phillip, W. A., The Future of Seawater Desalination: Energy, Technology, and the

Environment. Science 2011, 333 (6043), 712-717.



132



4. Elimelech, M.; Zhu, X.; Childress, A. E.; Hong, S., Role of membrane surface morphology in

colloidal fouling of cellulose acetate and composite aromatic polyamide reverse osmosis membranes. J. Membr. Sci. 1997, 127 (1), 101-109.

5. Zhu, X. Colloidal fouling of thin film composite and cellulose acetate reverse osmosis

membranes. 1996.

6. Zhu, X.; Hong, S.; Childress, A. E.; Elimelech, M. In Colloidal fouling of reverse osmosis

membranes: Experimental results, fouling mechanisms, and implications for water treatment, American

Water Works Association: 1995; pp 251-263.

7. Nielsen, W., Membrane Filtration and Related Molecular Separation Technologies. APV

Systems: 2000.

8. McGrath, J. E. In Chlorine resistant membranes for reverse osmosis and nanofiltration,

American Chemical Society: 2009; pp POLY-224.

9. McGrath, J. E.; Park, H. B.; Freeman, B. D. Chlorine resistant desalination membranes based on

directly sulfonated poly(arylene ether sulfone) copolymers. US20070163951A1, 2007.

10. Park, H. B.; Xie, W.; Freeman, B. D.; Paul, M.; Roy, A.; Sankir, M.; Lee, H.-S.; Riffle, J. S.; McGrath, J. E. In Chlorine-tolerant desalination membranes, American Chemical Society: 2008; pp

POLY-312.






13. Park, H. B.; Freeman, B. D.; Zhang, Z.-B.; Fan, G.-Y.; Sankir, M.; McGrath, J. E. In Water and salt transport behavior through hydrophilic-hydrophobic copolymer membranes and their relations to

reverse osmosis membrane performance, American Chemical Society: 2006; pp PMSE-495.


15. Rose, J. B. Sulfonated polyarylethersulphone copolymers. EP8894A1, 1980.

16. Rose, J. B. Sulfonated polyaryletherketones. EP8895A1, 1980.

17. Rose, J. B. Sulfonated poly(aryl ether ketones). EP41780A1, 1981.

133

4.0 Synthesis, Characterization and Post-sulfonation of a Polysulfone

Series Incorporating Hydroquinone for Reverse Osmosis Membranes

Ozma Lane, E. S. Jang, S. R. Choudhury, S. J. Mecham, B. D. Freeman, J. S. Riffle, J. E.

McGrath

4.1 Introduction

In recent decades, there has been a growing demand for production of potable drinking

water around the world. This growth is due to the combination of growing populations, as well as

contamination of freshwater sources and increasing energy costs for production of potable water.

This has have created a demand for more reverse osmosis plants and improved materials and

procedures.1 Of the methods available for the desalination of seawater, reverse osmosis has a clear

cost advantage over thermal processes that were developed decades earlier. Membrane separations

avoid the high cost of phase transitions.1 The predominant current material used in most reverse

osmosis applications is a crosslinked aromatic polyamide, produced by the Dow Chemical

Company under the trade name “FilmTec 30”. While this material has good transport properties,

it is susceptible to biofouling due to the rough surface produced during the interfacial crosslinking

reaction, as well as degradation of the polyamide in the presence of chlorine.2-4

Research has been underway to find novel materials or material modifications with

improved chlorine tolerance or resistance to biofouling.5-9 Disulfonated poly(arylene ether

sulfone)s are promising candidates, with excellent chlorine tolerance and a smooth material surface

that offers no rough topology for the adhesion of microbes.10, 11 Those polymers are prepared

directly from a disulfonated monomer that has two sulfonate groups on adjacent rings. However,

134

the sodium rejection in some of these materials is compromised in the presence of calcium.12 It

was hypothesized that this compromise may be a result of the close spacing between sulfonic acid

groups on adjacent rings on the polymer backbone.

In this research, we have investigated the development of a post-sulfonated material with

a monosulfonated hydroquinone unit as a potential candidate for reverse osmosis. Most previous

work on post-sulfonation of polysulfones required rather harsh conditions because the rings to be

sulfonated included both activated and deactivated rings toward the electrophilic aromatic

sulfonation reaction. Those studies on post-sulfonation of polysulfones found difficulties with the

control of molecular weight, targeting of sulfonation levels, and in the relative placement of

sulfonates along the chain (i.e., the microstructure).13 Alternatively, the hydroquinone-based

copolymers studied in this research can avoid these disadvantages by utilizing mild reaction

conditions.14, 15 The sulfonation proceeds only on the hydroquinone unit, while the other

comonomers feature electron-withdrawing groups that discourage side reactions. These materials

offer the advantage of being synthesized from commercially available reagents, allowing for

economical scale-up in addition to synthetic precision.

4.2 Experimental

4.2.1 Materials

Dichlorodiphenyl sulfone (DCDPS) was kindly provided by Solvay, and was recrystallized

in toluene and dried for 12 h under vacuum at 110 oC prior to use. Bisphenol sulfone (BisS), was

kindly provided by Solvay, and was recrystallized in methanol and dried for 12 h under vacuum

at 110 oC prior to use. Hydroquinone (HQ) was provided by Eastman Chemical Company and was

recrystallized in methanol and dried for 12 h under vacuum at 110 oC prior to use. m-Xylene,

135

sulfuric acid, and dimethylacetamide (DMAc) were obtained from Sigma Aldrich and used as

received. Sulfolane was obtained from Sigma Aldrich and heated with a 30% (v/v) portion of

toluene at 160 oC for 12 h to azeotropically remove any water. Potassium carbonate was obtained

from Sigma Aldrich and was dried for 12 h under vacuum at 180 oC prior to use.

4.2.2. Synthesis and Sulfonation

The hydroquinone sulfone (HQS) series was synthesized using a nucleophilic aromatic

substitution reaction as shown in Figure 4-1. A sample reaction follows: 4.955 g ( 45 mmol) of

HQ, 25.844 g (90 mmol) of DCDPS, 11.262 g (45 mmol) of BisS, and 100 mL of sulfolane were

added to a 3-necked round bottom flask equipped with nitrogen inlet, overhead stirrer, and

condenser with a Dean Stark trap. The reaction temperature was controlled with a thermocouple

in a salt bath. The reaction temperature was initially raised to 160 oC, 50 mL of m-xylene and 14.5

g (105 mmol) K2CO3 were added, and the reaction refluxed for 4 h to azeotropically remove any

water. The reaction temperature was then raised to 200-210 oC, and the m-xylene was removed

from the Dean Stark trap. After 36 h of reaction, the mixture was allowed to cool and diluted with

40 mL of DMAc. The solution was hot filtered to remove salts and precipitated in water. The

polymer was boiled with three changes of water to remove trace amounts of sulfolane, then dried

at 50 oC for 4 h, followed by 12 h under vacuum at 110 oC.

To sulfonate the polymer, the dried polymer powder was dissolved in a 10% solution of

sulfuric acid in a 3-necked round bottom flask equipped with nitrogen inlet and thermometer,

overhead stirrer, and condenser with a Dean Stark trap. An oil bath was used to maintain a reaction

temperature of approximately 50 oC. The reaction was stirred vigorously to promote rapid

dissolution and to break up any clumps of acid-swollen polymer. After 1 h of reaction, the solution

136

was precipitated into ice cold water, and rinsed thoroughly to remove all acid. Samples to be

analyzed were converted to their salt form by heating in 1.0 M NaCl for 2 h, then dried at 50 oC

for 4 h at atmospheric pressure, followed by 12 h under vacuum at 110 oC.

4.2.3. Characterization

4.2.3.1 NMR

Samples were dried overnight at 110 oC in a vacuum oven. In a scintillation vial with

molecular sieves, approximately 10 mg of the polymer was dissolved in 700 µL of d6-DMSO.

Spectra were collected on a Varian Unity Plus spectrometer operating at 400 MHz.

4.2.3.2. Water Uptake

Water uptake was measured gravimetrically. Polymer films (100 mg) were cast from a 10%

(w/v) solution of DMAc onto a clean glass plate and placed under an IR lamp for 6-12 h. Films

were removed via immersion in deionized water, and heated in water at 80 oC for 3 h to remove

any residual solvent. The samples were dried overnight at 110 °C under vacuum. Films were then

equilibrated in deionized water overnight, converted from the acid to their salt form by heating in

1.0 M NaCl at 80 oC for two h, then they were kept at room temperature in the water overnight.

They were blotted and weighed to obtain the wet weight (Wwet). Films were dried under vacuum

at 110 oC for 12 h in order to obtain dry weights (Wdry). The water uptake in the salt form was

determined as follows:



4.2.3.3. Molecular Weight Characterization

Size exclusion chromatography (SEC) was conducted on the polymers to measure

molecular weight distributions. The solvent was DMAc that was distilled from CaH2 and that

contained dry LiCl (0.10 M). The column set consisted of 3 Agilent PLgel 10-μm Mixed B-LS

137

columns 300x7.5 mm (polystyrene/divinylbenzene) connected in series with a guard column

having the same stationary phase. The column set was maintained at 50 °C. An isocratic pump

(Agilent 1260 infinity, Agilent Technologies) with an online degasser (Agilent 1260), autosampler

and column oven was used for mobile phase delivery and sample injection. A system of multiple

detectors connected in series was used for the analyses. A multi-angle laser light scattering

(MALLS) detector (DAWN-HELEOS II, Wyatt Technology Corp.), operating at a wavelength of

658 nm, a viscometer detector (Viscostar, Wyatt Technology Corp.), and a refractive index

detector operating at a wavelength of 658 nm (Optilab T-rEX, Wyatt Technology Corp.) provided

online results. The system was corrected for interdetector delay and band broadening. Data

acquisition and analysis were conducted using Astra 6 software from Wyatt Technology Corp.

Validation of the system was performed by monitoring the molar mass of a known molecular

weight polystyrene sample by light scattering. The accepted variance of the 21,000 g/mole

polystyrene standard was defined as 2 standard deviations (11.5% for Mn and 9% for Mw) derived

from a set of 34 runs.

4.2.3.4 Transport Property Measurement

The water permeability (L μm m-2

h-1

bar-1

or cm2

s-1

), salt permeability (cm2

s-1

), salt

rejection (%) and water/NaCl selectivity were determined at 25oC using stainless steel crossflow

cells. The pressure difference across the membrane (15.1 cm

2) was 400 psi. The aqueous feed

contained 2000 ppm NaCl, and the feed solution was circulated past the samples at a continuous

flow rate of 3.8 (L min-1

). The feed pH was adjusted to a range between 6.5 and 7.5 using a 10

g/L sodium bicarbonate solution. NaCl concentrations in the feed water and permeate were

138

measured with an Oakton 100 digital conductivity meter.

4.3 Results and Discussion

4.3.1 Synthesis

In the sulfonation reaction, the hydroquinone was selectively sulfonated while the electron-

withdrawing sulfone groups on all of the other rings did not react under the mild conditions utilized

(Figure 4-1).

Figure 4-1. Sulfonation of random poly(arylene ether sulfone) copolymers containing varied levels of hydroquinone.

The chemical compositions of the post-sulfonated polymers were investigated using 1H

NMR. In the representative spectrum shown below, the A and A1 peaks resonated at 7.95 and

7.83 ppm, the B and B1 peaks resonated at 7.24 and 6.97 ppm, and the hydroquinone peaks C, D,

and E resonated at 7.4, 7.15, and 7.05 ppm, respectively. The peak at 7.18 ppm corresponds to

unreacted hydroquinone, indicating an incomplete sulfonation reaction. The NMR procedure

described in Chapter 3 was used to determine the degree of sulfonation, and the IEC was calculated

from the NMR data.

139

Figure 4-2. Chemical structure and 1H NMR spectrum of SHQS-55 copolymer.

4.3.2 Membrane Properties

A series of three copolymers were synthesized with monomer ratios calculated to provide

a hydroquinone-containing repeat unit content of 45, 50, 55, and 60%. These values were targeted

to provide IECs of 0.99, 1.11, 1.21, and 1.33, respectively. The copolymer physical properties are

summarized in Table 4-1. Measured IECs and sulfonation levels were slightly lower than targeted

140

values, which is due to incomplete sulfonation as determined by the peak corresponding to

unreacted hydroquinone in Figure 4-2. This was due to an insufficient sulfonation reaction of 50oC

for one hour. The polymers had good molecular weights and produced clear, tough films. The

increase in molecular weight from SHQS-45 and -50 to the higher weights in SHQS-55 and -60

occurred with a new lot of sulfolane solvent, which may have contained fewer impurities than the

solvent used for previous synthetic reactions even after heating with toluene to azeotropically

remove any water. The solvent used initially also may have absorbed a large amount of water from

the atmosphere, and the toluene may have been insufficient to remove it entirely. Sulfolane and

water are highly miscible, and therefore even small amounts may be able to remain in the solvent

and interfere with achieving a high degree of polymerization.

Sample Target IEC IEC a Sulfonation b Water

Uptake (%) Mw

c (g/mole)

SHQS-45 0.99 0.92 42% 24 32,300

SHQS-50 1.11 1.06 48% 28 40,200

SHQS-55 1.21 1.14 51% 31 101,000

SHQS-60 1.33 1.21 54% 33 99,000

Table 4-1. Membrane Physical Properties

a) determined from 1H NMR

b) determined from 1H NMR

c) determined from SEC

4.3.3 Transport Data

The SHQS-50 membrane was evaluated for transport properties, and showed promising

sodium rejection and water permeability values (Table 4-2). The BPS-20 and BPS-30 materials

are a disulfonated poly(arylene ether sulfone) series that has been evaluated previously. The

141

number 20 in BPS 20 indicates that 20% of the repeat units had disulfonated comonomers, which

is similar in terms of IEC to a SHQS copolymer that contains 45-50% of the sulfonated

hydroquinone. Despite having a measured IEC only slightly higher than BPS20, the water

permeability was significantly higher, with only a small drop in sodium rejection. This may be due

to the monosulfonated repeat units forming fewer aggregates, allowing more homogenous

distribution of ionic domains throughout the membrane.

The mixed-feed data showed a constant salt rejection in the presence of up to 400 ppm of

calcium in the feed stream (Figure 4-2), with no screening by calcium of the Donnan exclusion

effect that is requisite to achieve high salt rejection. The compromised salt rejection of a 32%

disulfonated poly(arylene ether sulfone (BPS-32) is shown as a comparison. This confirms the

viability of post-sulfonated hydroquinone-containing copolymers as alternative candidates for

reverse osmosis membranes.

Table 4-2. Transport Properties of Selected Membranes

IECa (meq/g)

IECb

(meq/g) Water permeability

(L·μm/m2·h·bar)

NaCl rejection (%)*

SHQS-50 1.11 1.06 0.27 98.3

SHQS-60 1.21 1.14 0.45 98.3

BPS-20 0.93 1.01 0.05 99.2

BPS-30 1.34 1.32 0.96 91.3

aTheoretical calculation

bBased on 1H NMR

142

4.4 Conclusions

A series of poly(arylene ether sulfone)s with varying hydroquinone content were

synthesized and post-sulfonated. The IEC and sulfonation levels were slightly below the targeted

values, indicating incomplete sulfonation. This was confirmed by the small residual peak of

unreacted hydroquinone in the 1H NMR spectra. The post-sulfonated polymers produced tough,

ductile membranes with good molecular weights. Initial transport data shows promising

performance for the SHQS copolymer series, with transport properties comparable to those of the

disulfonated poly(arylene ether sulfone) systems. Mixed-feed testing confirms resistance to

0

2

4

6

8

10

12

0 50 100 150

Na

+ p

assag

e (

%)

Ca++

(ppm)

BPS-32

pSHQS-60

pSHQS-60 Al

Figure 4-3. Sodium rejection in post-sulfonated hydroquinone-containing copolymers

remains constant in the presence of calcium.

143

calcium-induced compromise of sodium rejection. Future work will include further optimization

of polymer composition in order to approach transport properties of state-of-the-art commercial

RO membranes.

References





3. Kucera, J. In Reverse osmosis membrane fouling control, CRC Press: 2010; pp 247-270.

4. Elimelech, M.; Zhu, X.; Childress, A. E.; Hong, S., Role of membrane surface morphology in

colloidal fouling of cellulose acetate and composite aromatic polyamide reverse osmosis membranes. J. Membr. Sci. 1997, 127 (1), 101-109.

5. Freeman, B.; McGrath, J. E. In Advances in polymer membranes for water purification, American

Chemical Society: 2010; pp MATNANO-191.

6. Guiver, M. D.; Robertson, G. P.; Tam, C. M., Functionalized polysulfones: methods for chemical

modification and membrane applications. Polym. Mater. Sci. Eng. 1997, 77, 347-348.

7. Lee, Y. T.; Kim, N. W.; Shin, D. H. Polyamide composite membrane having fouling resistance

and chlorine resistance and composite membrane fabrication. WO2010062016A2, 2010.

8. Meng, J.; Ni, L.; Li, X.; Zhang, Y. Antifouling chlorine-resistant polyamide RO composite

membrane and preparation method thereof. CN103331110A, 2013.

9. Xu, J.; Wang, Z.; Yu, L.; Wang, J.; Wang, S., Reverse osmosis membrane with regenerable anti -

biofouling and chlorine resistant properties. J. Membr. Sci. 2013, 435, 80-91.

10. Lee, C. H.; McCloskey, B. D.; Cook, J.; Lane, O.; Xie, W.; Freeman, B. D.; Lee, Y. M.;

McGrath, J. E., Disulfonated poly(arylene ether sulfone) random copolymer thin film composite

membrane fabricated using a benign solvent for reverse osmosis applications. Journal of Membrane Science 2012, 389 (0), 363-371.



12. Stevens, D. M.; Mickols, B.; Funk, C. V., Asymmetric reverse osmosis sulfonated poly(arylene

ether sulfone) copolymer membranes. J. Membr. Sci. 2014, 452, 193-202.


14. Lloyd, D. R.; Gerlowski, L. E.; Sunderland, C. D.; Wightman, J. P.; McGrath, J. E.; Igbal, M.;

Kang, Y., Poly(aryl ether) membranes for reverse osmosis. ACS Symp. Ser. 1981, 153 (Synth. Membr.,

Vol. 1), 327-50.

15. Bunn, A.; Rose, J. B., Sulfonation of poly(phenylene ether sulfones) containing hydroquinone

residues. Polymer 1993, 34 (5), 1114-16.

144

5.0 Synthesis and Characterization of a Hydrophilic-Hydrophobic

Multiblock Copolymer for Membrane Electrolysis Applications

Ozma Lanea, Jarrett Rowletta, Cortney Mittlesteadtb, Jason Willeyb, Amin Daryaeia, James

McGratha, Sue J. Mechama and J. S. Rifflea*

aMacromolecular Science and Engineering, Macromolecular and Interfaces Institute

Virginia Tech, Blacksburg, VA 24061

bGiner, Inc.

89 Rumford Avenue, Newton, MA 02466

*[email protected]

Abstract

Membrane electrolysis of water is a method of producing ultra high-purity hydrogen. Novel

materials are under investigation to design a membrane with high proton conductivity and minimal

hydrogen crossover. A multiblock hydrophilic-hydrophobic poly(arylene ether sulfone)

copolymer was synthesized and investigated. The focus was on the impact of film casting

conditions on membrane morphology and performance. Solvent selection and solution

concentration had a significant impact on long-range order and surface morphology of films. A

multiblock membrane with optimized casting conditions outperformed the conductivity/hydrogen

permeability ratio of Nafion™ by a factor of ~6. The hydrogen permeability of the partially

disulfonated poly(arylene ether sulfone) copolymer was much lower than for Nafion.

145

5.1 Introduction

The increasing energy demands created by a growing population and limited fossil fuel reserves

is driving research into alternative methods of energy production. Part of the diverse array of

energy sources includes renewable energy sources such as solar, hydroelectric and wind energy.

Renewable energy offers several advantages over fossil fuel-based resources that dominate the

current energy economy, a critical one being a reduction in the carbon dioxide output per unit of

energy. One of the criticisms of renewable energy is that it has unpredictable availability,

frequently generating power when not needed and being unavailable during hours of high demand.

A potential answer to both criticisms is the development of proton exchange membrane

(PEM) electrolysis systems, which can produce hydrogen via the electrolysis of water using energy

generated during off-peak hours.1 The hydrogen can be stored for use in vehicles which rely on

PEM fuel cells, or used in stationary power applications to supplement the electrical grid during

peak demand hours. PEM electrolysis has the advantage of producing very high purity hydrogen

gas, which is essential for use in PEM fuel cells.2 A good PEM for electrolysis should:

Have excellent mechanical properties to withstand the pressurized environment of the

electrolysis process

Have good proton conductivity

Have low hydrogen gas permeability

Be easily fabricated into a membrane

Resistant to chemical and oxidative degradation

Stable under a constant applied voltage

Have a high glass transition temperature

146

Be affordable3

Of these, two critical issues for efficiency and safety are high proton conductivity with low

hydrogen permeability. The latter is important to obtain high product purity and because the

crossover of hydrogen through the membrane to the oxygen-producing side becomes a safety

hazard.1, 4

The predominant commercial membrane used for PEM electrolysis is Nafion, a

poly(perfluorosulfonic acid) ionomer. While it provides good proton conductivity, its high

hydrogen permeability requires a thicker membrane to reduce crossover. Furthermore, its low

hydrated glass transition temperature hinders performance at temperatures above about 80 oC, with

reduced electrolysis performance and mechanical properties.5, 6 As a result, new approaches that

permit a higher operating temperature and pressure would significantly improve the performance

of PEM electrolysis operations.4

A number of multiblock copolymers have been explored as alternatives to Nafion for PEM

fuel cell applications.7-11 In this research, a hydrophilic-hydrophobic multiblock copolymer based

on partially disulfonated poly(arylene ether sulfone)s was synthesized and evaluated under

different casting conditions. Its conductivity and hydrogen permeability were evaluated as a

function of two different casting solvents. Bulk morphologies were investigated as a function of

casting solvent, solution concentration, and annealing conditions to develop morphology-

processing relationships for the membrane films.

5.2 Experimental

5.2.1 Materials

147

4,4’-biphenol (BP) was obtained from ChrisKev Company, Inc. and dried at 60 oC for 24

h under vacuum before use. Hydroquinone was kindly provided by Eastman Chemical, and was

recrystallized from ethanol and dried at 110 oC under vacuum prior to use. 4,4’-

Dichlorodiphenylsulfone (DCDPS) was kindly provided by Solvay Advanced Polymers and was

recrystallized from toluene and dried at 110 oC prior to use. 3,3’-Disulfonated-4,4’-

dichlorodiphenylsulfone (SDCDPS) was obtained from Akron Polymer Systems and was dried

under vacuum at 160 oC for 72 h before use. Hexafluoroisopropylidine diphenol (6FBPA) was

obtained from (Riedel-deHäen) and sublimed, then recrystallized from toluene, and dried under

vacuum at 110 oC for 12 h prior to use. N,N-Dimethylacetamide (DMAc) and N-methyl-2-

pyrrolidone (NMP) were obtained from Aldrich, and were vacuum-distilled from calcium hydride

onto molecular sieves and stored under nitrogen immediately before use. Potassium carbonate

(K2CO3, Aldrich) was dried under vacuum at 120 oC overnight before use. Toluene, cyclohexane,

methanol, acetone, and isopropyl alcohol (IPA) were obtained from Aldrich and used as received.

Concentrated sulfuric acid (H2SO4) was obtained from VWR and used to make a 0.5 M aqueous

solution. Difluorodiphenyl sulfone (DFBPS) was obtained from Aldrich and dried under vacuum

at 110 oC for 12 h prior to use.

Lead acetate was obtained from VWR and diluted to a 0.1% solution in deionized water

prior to use. Epo-fix™ epoxy was obtained from Electron Microscopy Services.

5.2.3 Synthesis

5.2.3.1 Synthesis of a HQSH100 Hydrophilic Block The hydrophilic block synthesis was adapted from a previously reported procedure.8 A

representative procedure follows: 8.41 g (15.75 mmol) of 98% SDCDPS, 1.93 g (17.53 mmol) of

HQ, and 35 mL of DMSO were added to a three neck flask equipped with an overhead stirrer, a

148

condenser with a Dean Stark trap, and nitrogen inlet. An oil bath equipped with a thermocouple

was used to heat the reaction to 135 oC, at which point toluene (55 mL) and potassium carbonate

(2.78 g, 20.1 mmol) were added and the mixture was refluxed for 4 h to azeotropically remove any

water. The reaction was then heated to 150 oC, and the toluene was removed via the Dean Stark

trap. After 96 h of reaction, the solution was diluted with 30 mL of DMF and hot filtered to remove

salts, then precipitated in isopropanol. A second hydrophilic block was produced with a 4,000

g/mol number average molecular weight.

5.2.3.2. Synthesis and Endcapping of Hydrophobic Blocks

DCDPS (9.54 g, 33.22 mmol), 6FBPA (12.04 g, 35.81 mmol) and DMAc (100 mL) were

added to a three neck flask equipped with an overhead stirrer, a condenser with Dean Stark trap,

and a nitrogen inlet. An oil bath equipped with a thermocouple was used to heat the reaction to

140 oC, at which point toluene (55 mL) and potassium carbonate (5.70 g, 41.2 mmol) were added

and the mixture was refluxed for 4 h to azeotropically remove any water. The reaction was then

heated to 160 oC, and the toluene was removed via the Dean Stark trap. The reaction proceeded

for 24 h, then the temperature was lowered to 120 oC and decafluorobiphenyl (34.61 g, 103.6

mmol) was added (20-fold molar excess). The endcapping reaction proceeded for 8 h. The solution

was hot filtered to remove salt, and precipitated in isopropanol. The yield was 81%.

5.2.3.3 Coupling Reaction of the Hydrophilic and Hydrophobic Blocks

HQS100 (4.62 g) and DMF (45 mL) were added to a three-neck, 100-mL round bottom

flask equipped with a mechanical stirrer, condenser, nitrogen inlet and Dean-Stark trap. The

solution was heated to 130 oC, then cyclohexane (15 mL) and K2CO3 (0.2 g, 1.45 mmol) were

added and the mixture was refluxed for 4 h to remove water from the system. The cyclohexane

was drained from the Dean Stark trap and then the mixture was cooled to 90oC. After cooling, the

6FBPS0 hydrophobic oligomer (4.25 g) was added. The bath temperature was raised to 120oC and

149

maintained for 24 h. The reaction mixture was diluted with DMF (35 mL) and allowed to cool to

room temperature. The multiblock copolymer was precipitated into isopropanol (1000 mL) and

stirred for 12 h. The product was filtered, then washed in deionized (DI) water at 90oC for 12 h,

filtered again, and then dried in vacuo at 150oC for 24 h.

5.2.4.1 Synthesis of the Fluorine-Terminated Hydrophobic Block

An additional hydrophobic block was synthesized with DFBPS (12.18 g, 47.9 mmol),

6FBPA (15.24 g, 45.4 mmol) and NMP (135 mL) were added to a three neck flask equipped with

an overhead stirrer, a condenser with Dean Stark trap, and a nitrogen inlet. An oil bath equipped

with a thermocouple was used to heat the reaction to 148 oC, at which point cyclohexane (70

mL) and potassium carbonate (7.3 g, 52.8 mmol) were added and the mixture was refluxed for 4

h to azeotropically remove any water. The reaction was then heated to 165 oC, and the

cyclohexane was removed via the Dean Stark trap. The reaction proceeded for 12 h. The solution

was diluted with NMP (50 mL), hot filtered to remove salt, and precipitated in isopropanol.

5.2.4.2 Coupling of the Hydrophilic and Fluorine-Terminated Hydrophobic Blocks

HQS100 (16.4 g, 2.83 mmol) and DMF (190 mL) were added to a three-neck, 100-mL

round bottom flask equipped with a mechanical stirrer, condenser, nitrogen inlet and Dean-Stark

trap. The solution was heated to 130 oC, then cyclohexane (60 mL) and K2CO3 (1.0 g, 7.25

mmol) were added and the mixture was refluxed for 4 h to remove water from the system. The

cyclohexane was drained from the reaction and then the mixture was cooled to 90oC. After

cooling, the 6FBPS0 hydrophobic oligomer (21.5 g, 2.15 mmol) (~ 10,000 g/mol) was added.

Excess hydrophilic monomer was used in this case due to the large difference in molecular

weight between the two oligomers. The bath temperature was raised to 145oC and kept at this

temperature for 48 h. The reaction mixture was diluted with DMF and allowed to cool to room

temperature. The multiblock copolymer was precipitated into isopropanol (2000 mL) and stirred

150

for 12 h. The product was filtered then washed in deionized (DI) water at 90oC for 12 h, filtered

again, and then dried in vacuo at 150oC for 24 h.

5.2.5 Characterization

5.2.5.1 NMR spectroscopy

1H and 19F NMR spectroscopy analyses were performed on a Varian INOVA 400

spectrometer operating at 400 and 376 MHz respectively. Spectra were obtained from a 10%

solution (w/v) solution in d-CDCl3 at ambient temperature.

5.2.5.2 Film Casting, Annealing, and Acidification

Films were cast by dissolving the copolymer to afford either 5% or 10% (w/v) solutions in

DMAc or NMP. The solutions were filtered through a 0.1 m PTFE Acrodisc filter onto a clean

glass plate. The films were heated under an IR lamp for 12 h at 60 oC, and then under vacuum at

110oC for 24 h. Annealed films were produced by additionally heating under vacuum at 220 oC

for 12 h. Films were acidified by boiling in 0.05 M H2SO4 for 2 h, and then in deionized water for

2 h. Note that for cases where the polymer was annealed, it was acidified after the annealing step.

5.2.3.3 Water Uptake

Water uptake was measured gravimetrically. Polymer films (100 mg) in acid form were

equilibrated in deionized water overnight, then kept at room temperature for 12 h. They were

blotted with a Kimwipe to remove surface water and weighed to obtain the wet weight (Wwet).

Films were dried under vacuum at 110 oC for 12 h in order to obtain dry weights (Wdry). The water

uptake in the salt form was determined as follows:



151

5.2.3.3 Transmission Electron Microscopy

For transmission electron microscopy imaging, acidified membranes were stained with a

0.1% lead acetate aqueous solution to enhance the electron density of the hydrophilic block and

provide contrast within the sample. Samples were sputter coated with a Au/Pd alloy to improve

adhesion, then embedded in epoxy Epo-fix™ (electron microscopy services) and microtomed into

approximately 70-nm thick sections with a diamond knife. Samples were then imaged on a JEOL

2100 Transmission Electron Microscope using an accelerating voltage of 120 kV.

5.3 Results

5.3.1 Synthesis

5.3.1.1 Hydrophilic Oligomer Synthesis

The synthesis of the hydrophilic HQS100 block via nucleophilic aromatic substitution is

illustrated in Figure 5-1. The long reaction time was necessary due to the relatively low reactivity

of both monomers. DMSO was used as a reaction solvent due to limited solubility of both

monomers in other polar aprotic solvents.

Figure 5-1. Schematic of hydrophilic oligomer synthesis.

152

1H NMR was utilized to analyze the structure of the resulting oligomer (Figure 5-2). The

protons on the sulfonated aromatic rings are attributed to the indicated A, B, and C protons located

at 8.25, 7.83, and 6.93 ppm, respectively. The main chain hydroquinone protons (D) resonate at

7.08 ppm. The phenoxide endgroups resonate at 6.75 and 6.85 ppm (E and F). The singlet peak at

7.9 ppm is attributed to a small amount of dimethylformamide contamination. The peaks at 8.32,

7.7 and 7.65 ppm are attributed to SDCDPS endgroups (A’, B’ and C’). While Figure 5-2 illustrates

both types of endgroups, it should be noted that they are not present in a 1:1 ratio.

Integrations of the endgroup peaks show a ratio of approximately 60% SDCDPS and 40%

HQ endgroups. The molecular weight calculated from the ratio of endgroup peaks to the (A) proton

on the SDCDPS repeat unit was approximately 3000 g/mol. The initial monomer ratios were

calculated to produce an oligomer with phenoxy endgroups. However, due to the slow reactivities

of both monomers, both types of endgroups were present after reaction.

153

Figure 5-2. 1H NMR of a 3K hydrophilic block

5.3.1.2 Synthesis of the Hydrophobic Oligomer

The hydrophobic oligomer was synthesized from the reaction of 6FBPA with DCDPS, and

then it was endcapped with a large excess of decafluorobiphenyl (Figure 5-3). Decafluorobiphenyl

was used as an endcapping reagent in order to increase reactivity, so that the block copolymer

coupling reaction could be carried out at a sufficiently low temperature to avoid transetherification.

154

Transetherification is a major concern in the synthesis of poly(arylene ether sulfone) multiblocks

as the interchange of ether groups would disrupt the block structure.

Figure 5-3. Synthesis and endcapping of 6FS hydrophobic block.

1H and 19F NMR were used to confirm the expected structure of the resulting oligomer

(Figures 5-4 and 5-5). The aromatic protons from the 6FBPA unit resonated at 7.35 (B) and 7. 15

(A) ppm, while the aromatic protons from the sulfone unit resonated at 7.15 (C) and 7.92 ppm (D).

In the 19F NMR, the peak at -64 resonates with the CF3 groups on the oligomer backbone. The -

141, -150, -152, and -160 resonate with the C, A, D, B fluorines indicated in Figure 5-5. The ratio

of the integrals of the 6F fluorines and the C fluorines on the endgroups was used to determine a

number average molecular weight of 10,000 g/mol.

For the fluorine-terminated hydrophobic block, the molecular weight was determined by

19F NMR to be 6100 g/mol.

155

Figure 5-4. Structure and 1H NMR spectrum of DFBP-endcapped 6FS hydrophobic oligomer.

156

Figure 5-5. 19F NMR of DFBP-endcapped hydrophobic 6FS oligomer.

5.3.1.3 Coupling of Hydrophilic and Hydrophobic Oligomers

102.93

8.00

2.01

8.00

8.00

3.71

3.94

157

Figure 5-6. Coupling reaction to form the multiblock copolymer.

158

Figure 5-7. 1H NMR of the multiblock copolymer.

The block copolymers were prepared by the coupling reaction of the fluorine terminated

hydrophobic oligomer and the phenoxide endgroups of the hydrophilic oligomer (Figure 5-6).

Oligomers terminating in chlorine endgroups at either end were unreactive under the reaction

conditions, and oligomers with both a phenoxide and chlorine endgroup likely formed a terminal

block of the multiblock copolymer.

The A, B, and C protons of the disulfonated unit in Figure 5-7 resonate at 8.22, 7.81, and

6.94 ppm, respectively. The A’ peak corresponding to the SDCDPS endgroup resonates at 8.32.

The ratio of A and A’ integrals indicates that about 40% of the SDCDPS endgroups were lost in

the coupling reaction or during isolation of the block copolymer. The D peak corresponding to the

hydroquinone protons resonates at 7.07 ppm, and the E peak corresponding to the inner protons

159

on the DCDPS unit resonates at 7.94 ppm. The F and G peaks of the ether-proximate protons on

both the 6FBPA and DCDPS units overlap, resonating at approximately 7.17 ppm. The H peak

corresponding to the inner protons on the 6FBPA unit resonates at 7.36 ppm.

Integrations of the A, A’ and H peaks were used to calculate the ion exchange capacity

(meq of ions/g), listed with other membrane properties in Table 5-1. The IEC of the 10K-3K

hydrophobic-hydrophilic multiblock copolymer was determined to be 1.10 meq/g. 1H NMR

analysis indicated a hydrophilic weight fraction of 35%, and a hydrophobic weight fraction of

65%. This is slightly higher than that of the Nafion 1100 films (0.91 meq/g) which were also

evaluated for proton conductivity and hydrogen permeability. The second multiblock, coupled

from a fluorine-terminated hydrophobic block without a decafluorobiphenyl linkage group, the

hydrophobic oligomer was 6100 g/mol and the hydrophilic was 4000 g/mole. The overall

composition was determined to be 60% hydrophobic and 40% hydrophilic. Despite having very

different IECs, the samples showed very similar water uptake behavior. This may be due to

differing morphologies as a result of the distinct oligomer block compositions.

Hydrophobic-

Hydrophilic block

length

Ion Exchange

Capacity (NMR)

Water Uptake (%)

10K-3K 1.10 47

6K-4K 1.41 48

Table 5-1. Membrane Properties of Multiblock Copolymers

5.3.1.4 Bulk Morphology

160

5.3.1.5 Effect of Annealing and Molecular Weight

Figure 5-4. TEM image of a cross-section of the 11K-3K 6FBPS0-HQSH100 copolymer, (left) before and (right) after

annealing. All films were cast from 5% DMAc. Arrow points across the film thickness (glass side to air side of film).

Figure 5-8 shows the TEM images of through-plane cross-sections of thin films of

multiblock copolymers. Dark areas indicate hydrophilic regions of the copolymer films that have

been stained with lead acetate, while light areas indicate hydrophobic regions. The annealed

samples were heated for 12 hours at 220 oC under vacuum, which is just above the Tg of the

hydrophobic block. The impact of annealing on long-range order and lamellar orientation is shown

in the 11K-3K 6FBPS0-HQSH100 copolymers cast from 5% DMAc. The images were obtained

under ultra high vacuum, and so during PEM electrolysis, the dark hydrophilic regions would be

much larger. For the 8K-5K 6FBPS0-HQSH100 copolymer, the upper left-hand image illustrates

the non-annealed sample, which shows little long-range assembly, and no apparent orientation of

morphology with respect to any direction. However, after annealing between the two Tgs of the

hydrophilic and hydrophobic phases, the same material shows both long range ordered lamellae,

persisting in some cases for over 100 nm, and with significant orientation of the lamellae with

161

respect to the membrane surfaces. This orientation is likely due to preferential affinity of the

hydrophobic block for the membrane surfaces, due to lower surface energy. During the annealing

process, the hydrophobic block forms lamellae parallel to the surface, which propagates a degree

of orientation throughout the membrane thickness as the hydrophobic domains re-assemble. The

higher block length copolymer shows more long range order before annealing (lower left and lower

right images), but also shows orientation parallel to the surfaces post-annealing. In both cases, the

morphological comparisons were made from sections of the same film, one stained after casting

and acidification and one stained after casting, annealing and acidification, to avoid any variations

between different films.

5.3.1.6 Impact of Casting Solution on Film Morphology

A comparison was also made between films cast from 5% solutions in NMP (left) and

DMAc (right) from the 8K-5K 6FBPS0-HQSH100 copolymer (Figure 5-9). The films were cast,

annealed, and acidified together to ensure identical casting conditions. Even though the solvent

should be almost entirely removed prior to annealing, it appears there was still a solvent effect in

terms of the capacity for the morphology to become re-ordered after annealing. The NMP-cast

film had a less ordered lamellar morphology, with no apparent orientation, while the DMAc-cast

film again showed an ordered, oriented lamellar structure.

162

5.3.1.7. Conductivity and Permeability

The conductivity data for the copolymers compares two different casting methods: the 10K-3K

copolymer cast from 10% polymer in NMP, and the copolymer cast from 5% polymer in DMAc.

In these results, the film cast from DMAc shows a significant increase in proton conductivity over

the NMP-cast film, but also a decrease in hydrogen (H2) permeability. These two improvements

combine to provide a sharp increase in the conductivity/permeability ratio for the NMP-cast film

relative to the DMAc-cast film. The improvement in proton conductivity is likely due to the

improvements in long-range order of the hydrophilic block providing longer proton-conducting

channels. However, the increase in orientation of the lamellae parallel with the plane of the film

likely drives the drop in hydrogen permeability. The orientation increases the number of layers

that the hydrogen must cross in order to permeate the thickness of the membrane, and each lamella

represents an additional step of adsorption, diffusion and desorption into the next layer. Thus it is

1 0 0 n m1 0 0 n m

Figure 5-5. TEM images of the 10K-4K 6FS-HQSH100 copolymer cast from (left) a 5% NMP solution and (right) a 5% DMAc solution. Arrow again indicates thickness of film, going from glass side of film to air side.

163

hypothesized that the lamellae effectively form hurdles that slow the hydrogen transport through

the thickness of the film.

Table 5-1. Proton Conductivity and hydrogen gas permeability, all numbers relative to Nafion 100.

Conductivity Permeability C/P Ratio

Giner HQSH100-

6FBP0 0.6 0.5 1.1

VT HQSH100-

6FBPS0

1.1 0.18 6.1

Nafion 1100 1.0 1.0 1.0

5.4 Conclusions

A hydrophobic-hydrophilic 10K-3K multiblock poly(arylene ether) sulfone copolymer was

synthesized with terminal hydrophilic blocks. Thin films cast from 5% polymer in DMAc and

then annealed at 220 oC (above Tg of hydrophobic block) had highly aligned lamellar morphology

parallel to the plane of the film. Proton conductivity in liquid water at 95 oC exceeded that of

Nafion 1100, with significantly lower hydrogen (H2) permeability. This is important for efficiency

of membrane electrolysis of water, because high conductivities with low hydrogen cross-over are

a priority for alternative materials development.

Morphology was characterized as a function of casting solution, concentration, and

acidification treatment. Annealing was shown to not only increase long-range order as expected,

but also induced orientation parallel to the film surfaces which propagated throughout the bulk of

164

the film. The high weight fraction of the hydrophobic oligomer together with highly mobile,

hydrophilic endgroups in water led to materials with ordered surfaces and orientation of the

lamellae in the plane of the film as illustrated in the TEM micrographs. The small weight fraction

of the hydrophilic phase combined with the greater mobility of the hydrophilic end blocks likely

contributed to the high proton conductivity combined with the low hydrogen permeability.

1. Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D., A comprehensive review on PEM water

electrolysis. Int. J. Hydrogen Energy 2013, 38 (12), 4901-4934.

2. Smolinka, T.; Rau, S.; Hebling, C. In Polymer electrolyte membrane (PEM) water electrolysis,

Wiley-VCH Verlag GmbH & Co. KGaA: 2010; pp 271-289.

3. Arico, A. S.; Siracusano, S.; Briguglio, N.; Baglio, V.; Blasi, A.; Antonucci, V., Polymer

electrolyte membrane water electrolysis: status of technologies and potential applications in combination

with renewable power sources. J. Appl. Electrochem. 2013, 43 (2), 107-118.

4. Gandia, L. M., Arzamendi, G., Diéguez, P.M., Renewable hydrogen technologies: production,

purification, storage, applications and safety. Elsevier

2013.

5. Mauritz, K. A.; Moore, R. B., State of understanding of Nafion. Chem. Rev. (Washington, DC, U.

S.) 2004, 104 (10), 4535-4585.

6. Phillips, A. K.; Moore, R. B., Mechanical and transport property modifications of

perfluorosulfonate ionomer membranes prepared with mixed organic and inorganic counterions. J. Polym.

Sci., Part B: Polym. Phys. 2006, 44 (16), 2267-2277.

7. Chen, Y.; Rowlett, J. R.; Lee, C. H.; Lane, O. R.; Van, H. D. J.; Zhang, M.; Moore, R. B.;

McGrath, J. E., Synthesis and characterization of multiblock partially fluorinated hydrophobic

poly(arylene ether sulfone)-hydrophilic disulfonated poly(arylene ether sulfone) copolymers for proton

exchange membranes. J. Polym. Sci., Part A: Polym. Chem. 2013, 51 (10), 2301-2310.

8. Lee, H.-S.; Lane, O.; McGrath, J. E., Development of multiblock copolymers with novel

hydroquinone-based hydrophilic blocks for proton exchange membrane (PEM) applications. J. Power

Sources 2010, 195 (7), 1772-1778.

9. Lee, H.-S.; Roy, A.; Lane, O.; Dunn, S.; McGrath, J. E., Hydrophilic-hydrophobic multiblock

copolymers based on poly(arylene ether sulfone) via low-temperature coupling reactions for proton

exchange membrane fuel cells. Polymer 2008, 49 (3), 715-723.

10. Rowlett, J. R.; Chen, Y.; Shaver, A. T.; Lane, O.; Mittelsteadt, C.; Xu, H.; Zhang, M.; Moore, R.

B.; Mecham, S.; McGrath, J. E., Multiblock poly(arylene ether nitrile) disulfonated poly(arylene ether

sulfone) copolymers for proton exchange membranes: Part 1 synthesis and characterization. Polymer

2013, 54 (23), 6305-6313. 11. Rowlett, J. R.; Shaver, A. T.; Lane, O.; Mittelsteadt, C.; Xu, H.; Zhang, M.; Moore, R. B.;

Mecham, S.; McGrath, J. E. In Properties and performance of multiblock copolymers based upon a

varying hydrophilicity backbone, American Chemical Society: 2014; pp ENFL-76.

166

6.0 Synthesis and Characterization of Hydrophobic-Hydrophilic Block

Copolymers for Proton Exchange Membranes

Ozma R. Lane, Rachael A. VanHouten, Desmond J. VanHouten, James E. McGrath, J. S. Riffle*

Macromolecular Science and Engineering, Macromolecular and Interfaces Institute

Virginia Tech, Blacksburg, VA 24061

*[email protected]

Submitted to the Journal of Membrane Science, August 2015

Abstract

A series of hydrophobic-hydrophilic multiblock poly(arylene ether sulfone) copolymers based on

highly fluorinated and sulfonated monomers were synthesized and characterized for use as proton

exchange membranes in fuel cell applications. A preformed hydrophilic oligomer was reacted

with the monomers for the hydrophobic component to form segmented copolymers containing

highly fluorinated hydrophobic segments and 100% disulfonated hydrophilic blocks via a

nucleophilic aromatic substitution step polymerization reaction. This approach afforded high

molecular weight, transparent, and ductile copolymers with good mechanical properties. At

comparable ion exchange capacities, water uptake increased with block length, suggesting that the

extent of nanophase separation was a function of block length. This was confirmed by TEM and

AFM images. These copolymers showed improved proton conductivity under partially-

humidified conditions with increasing block length. A series of multiblock hydrophobic-

167

hydrophilic copolymers using the same co-monomers and oligomer molecular weights was also

synthesized and evaluated for comparison.

6.1 Introduction

With the growing economic and environmental costs associated with fossil fuel extraction,

there is an increasing need for viable alternatives to current combustion engine technology.

Hydrogen-powered proton exchange membrane fuel cells (PEMFCs) are viewed as a promising

candidate for powering 21st century vehicles.1, 2 Fuel cells using other fuels—for example,

methanol or ethanol—may also be adapted for portable power applications such as personal

electronic devices (e.g., laptop computers and cell phones). PEMFCs have a number of advantages

compared to other types of fuel cells.1 For example, they are able to generate more power for a

given volume or weight of the fuel cell. This high power density enables their use in applications

such as personal vehicles. In addition, their low operating temperature (below 100ºC) allows rapid

start-up. Most importantly, PEMFCs have a negligible environmental impact with only water and

heat produced during operation. These traits make them superior candidates for automotive power

applications.3

There are, however, a number of technical hurdles to overcome before the large-scale

commercial production of PEMFC vehicles can be realized. In addition to the difficulty of

hydrogen storage, hydrogen distribution and delivery will require the cons truction of a

costly infrastructure combining transport and storage of hydrogen and the development

of a dense network of refueling stations.4-6 But as those problems are addressed, research

must also resolve the challenges associated with materials development. Specifically, the

predominant commercial membrane in use today is Nafion®, which is a poly(perfluorosulfonic

168

acid) ionomer. While it displays good proton conductivity, it has several drawbacks which at

present limit practical commercial application. These include high permeability to both hydrogen

and methanol, excessive materials cost, the difficulty in processing the copolymer, and a low

operating temperature of about 80oC under hydrated conditions.7, 8 A number of alternate proton

exchange membrane candidates, membrane treatment procedures, and membrane additives are

under development to overcome these challenges.9-16

Our group has investigated partially disulfonated statistical and multiblock poly(arylene

ether sulfone) copolymers as alternative PEM candidates. These materials offer improvement both

in raw materials cost and ceiling operating temperature, but the statistical copolymers have reduced

conductivities as humidity is decreased.17, 18 The later development of nanophase-separated

multiblock copolymers has shown improved proton conductivity under partially humidified

conditions.19-21 The formation of hydrophilic channels that retain water improves conductivity,

while the highly hydrophobic channels provide mechanical strength and reduce water loss. The

multiblock copolymers allow for higher ion exchange capacities (IEC) without loss of mechanical

properties relative to the statistical copolymers. The nanophase separation additionally results in

anisotropic water uptake behavior which could reduce membrane stresses due to reduced in-plane

swelling. 19-21 During fuel cell operation, stresses are caused by in-plane swelling and deswelling

due to hydration and dehydration of the membrane electrode assembly (MEA). This can cause

failure between the electrode and polymer electrolyte layers. The swelling behavior becomes more

anisotropic with higher oligomer block lengths. Thus, there is interest in developing nanophase-

separated multiblock copolymers to optimize PEMFC performance both in terms of proton

conductivity and fuel cell durability.16, 22

169

Previously in our group, highly sulfonated, highly fluorinated multiblock copolymers were

synthesized from a hydrophobic block comprised of decafluorobiphenyl and bis(4-hydroxyphenyl)

sulfone coupled with a hydrophilic block of 4,4’-biphenol and SDCDPS.23 Those copolymers

showed good proton conductivity and mechanical properties. A highly fluorinated monomer was

chosen to be incorporated into the hydrophobic oligomer to achieve phase separation. The highly

activated decafluorobiphenyl endgroups were also an advantage so that the oligomer coupling

reaction could be conducted at relatively low temperatures to avoid any transetherification. A

series of multiblock copolymers with these structures having approximately equal block lengths

of the oligomers, but with varied block lengths were investigated. These multiblock copolymers

with the longer block lengths featured improved conductivities under partially humidified

conditions, comparable or competitive with that of Nafion® 212.

The focus of this research has been to simplify the method for making the multiblock

copolymers. We compared the properties of a nanophase-separated series of copolymers with

analogous structures that were prepared in a one-pot synthetic process versus the more traditional

“two preformed” oligomer approach. In this technique, the hydrophilic block was preformed with

phenol endgroups and then this was copolymerized with the monomers for the hydrophobic block

in-situ. While this general technique is not new, it was not known whether this simplified approach

would yield materials with the excellent transport properties that are ideal for the fuel cell

application. This method is an attractive synthetic technique for several reasons. Using this one-

pot technique, multiblock copolymers can be synthesized more easily in a shorter amount of time

because there is no need to synthesize both oligomers separately and then couple them together.

The one-pot “segmented” technique can also sometimes allow for more choice of solvents

compared to coupling two preformed oligomers that may differ greatly in chemical polarity. This

170

avoids the often observed polymer-polymer incompatibilities between two oligomers that can

complicate block copolymer synthesis.

In this paper, we discuss the synthesis and properties of multiblock copolymers comprised

of decafluorobiphenyl, bis(4-hydroxyphenyl)sulfone, biphenol and SDCDPS using two synthetic

routes. In the oligomer-oligomer method, each oligomer was separately preformed, then they were

coupled. This was compared with a segmented, streamlined approach wherein only the hydrophilic

oligomer was preformed, then that oligomer was copolymerized with the monomers to form the

hydrophobic block in-situ. The effect of block length on copolymer properties including proton

conductivity, water uptake, and dimensional swelling are discussed.

6.2 Experimental Section

6.2.1 Materials

Decafluorobiphenyl (DFBP) was obtained from Matrix Scientific and dried under vacuum

at ambient temperature for 12 h. Bis(4-hydroxyphenyl)sulfone (Bis-S) was kindly supplied by

Solvay Advanced Polymers and dried under vacuum at 60 oC for 24 h before use. Monomer grade

4,4’-biphenol (BP) was obtained from ChrisKev Company, Inc. and dried at 60 oC for 24 h under

vacuum before use. 4,4’-Dichlorodiphenylsulfone (DCDPS) was kindly provided by Solvay

Advanced Polymers and used as received to synthesize 3,3’-disulfonated-4,4’-

dichlorodiphenylsulfone (SDCDPS) according to a previously reported procedure.24 SDCDPS

was dried under vacuum at 160 oC for 48 h before use. N,N-Dimethylacetamide (DMAc, Aldrich)

and N-methyl-2-pyrrolidone (NMP, Aldrich) were vacuum-distilled from calcium hydride onto

171

molecular sieves and stored under nitrogen immediately before use. Potassium carbonate (K2CO3,

Aldrich) was dried under vacuum at 120 oC overnight before use. Toluene, cyclohexane, acetone,

and isopropyl alcohol (IPA) were obtained from Aldrich and used as received. Concentrated

sulfuric acid (H2SO4) was obtained from VWR and used to make a 0.5 M aqueous solution.

6.2.2. Synthesis of Phenoxide-Terminated Hydrophilic Blocks (BPS-100).

Phenoxide-terminated hydrophilic blocks were synthesized using a previously published

procedure. The targeted molecular weights of the blocks ranged from 3000 to 9000 g/mol. In a

typical procedure to achieve a Mn of 3000 g/mol, the following conditions were utilized. BP

(4.7322 g, 25.41 mmol), SDCDPS (10.2764 g, 20.92 mmol), and DMAc (75 mL) were charged to

a three-neck, round-bottom flask, equipped with a mechanical stirrer, a Dean-Stark trap and

condenser, and N2 inlet. The reaction temperature was maintained at 85 oC utilizing a

thermocouple-controlled oil bath and the monomers were dissolved. K2CO3 (4.039 g, 29.23 mmol)

and toluene (38 mL) were added. The temperature of the oil bath was increased to 155 oC, and the

reaction was refluxed for 4 h to azeotropically remove water. Toluene was removed from the

system by increasing the bath temperature to 175 oC. The reaction was allowed to proceed for 96

h. After cooling to ambient temperature, the solution was filtered using an aspirator to remove

salts and precipitated into acetone. The resulting oligomer was dried at 110 oC for at least 24 h

under vacuum and had an Mn of 3300 g/mol determined by end group analysis using 1H NMR.

6.2.3. Synthesis of Segmented Copolymers.

This copolymer series is referred to BisSF-BPSH100, where BisSF describes to the

hydrophobic block synthesized from bis(4-hydroxyphenyl) sulfone and decafluorobiphenyl, and

BPSH100 refers to the hydrophilic block synthesized from SDCDPS and 4,4’-biphenol. A sample

172

copolymerization procedure was as follows. A three-neck, round-bottom flask, equipped with a

mechanical stirrer, a Dean-Stark trap, and a condenser, and N2 inlet was loaded with BPS-100

(Mn=3300 g/mol, 4.0418 g, 1.225 mmol), Bis-S (1.6700 g, 6.673 mmol), and NMP (29 mL). After

dissolution of the reactants, K2CO3 (1.255 g, 9.081 mmol) and cyclohexane (5 mL) were added to

the reaction solution. The oil bath was heated to 110 oC, and the reaction was azeotroped to remove

water for 4 h. The cyclohexane was drained, and the oil bath temperature was lowered to 90 oC.

DFBP (2.6391 g, 7.899 mmol) and NMP (13 mL) were added to the reaction flask and the

temperature was maintained at 90 oC for 36 h. The viscous solution was cooled, filtered with an

aspirator, and precipitated into IPA (1 L). The product was filtered and washed in deionized water

at 60 oC for 12 h and acetone for 12 h. It was dried under vacuum at 110 oC for 24 h before casting

into films.

6.2.4. Synthesis of BisSF-BPSH100 Multiblock Copolymer Controls.

Multiblock copolymer controls were synthesized according to a previously published

method.23 Some modifications were made to the procedure and are reflected below.

6.2.4.1. Synthesis of Fluorine-Terminated Hydrophobic Blocks (BisSF) (2).

Fluorine-terminated hydrophobic blocks were synthesized with targeted molecular weights

ranging from 3000 to 9000 g/mol. For a Mn of 3000 g/mol, the following synthetic procedure was

utilized. Bis-S (2.5045 g, 10.01 mmol), DFBP (4.0263 g, 12.05 mmol), and DMAc (38 mL) were

added to a three-neck, round-bottom flask, equipped with mechanical stirrer, Dean-Stark trap with

condenser, and N2 inlet. The oil bath was heated to 50 oC. After dissolving the monomers, K2CO3

(1.798 g, 13.01 mmol) and cyclohexane (7 mL) were added. The bath temperature was gradually

raised to 110 oC over 30 min. The reaction was allowed to proceed for 5 h at 110 oC. After cooling

to ambient temperature, the reaction was filtered to remove salts and precipitated into a solution

173

of methanol:water (1:1 v:v, 1 L). The oligomer was washed for 12 h in DI water and dried at 90

oC for 24 h under vacuum before further use. It had a Mn of 3200 g/mol determined by endgroup

analysis using 19F NMR.

6.2.4.2. Synthesis of Phenoxide-Terminated Hydrophilic Blocks (BPS-100) (1).

Phenoxide-terminated hydrophilic blocks were synthesized with targeted molecular weights

ranging from 3000 to 9000 g/mol. To obtain a targeted molecular weight of 3000 g/mol, the

following reaction procedure was utilized. BP (0.7939 g, 4.263 mmol), SDCDPS (1.7466 g, 3.555

mmol), and NMP (20 mL) were added to a three-neck, round-bottom flask, equipped with a

mechanical stirrer, Dean-Stark trap with condenser, and N2 inlet. The oil bath was heated to 85

oC. Upon dissolution of the monomers, K2CO3 (0.766 g, 5.543 mmol) and toluene (10 mL) were

added to the flask. The bath temperature was increased to 155 oC, and the reaction was refluxed

for 4 h. Toluene and water was removed azeotropically, then the reaction bath temperature was

increased to 190 oC for 36 h. The bath temperature was lowered to 85 oC. The resulting oligomer

had a Mn of 3300 g/mol determined by end group analysis using 1H NMR. This product was not

isolated and was used directly in the synthesis of a BisSF-BPS100 multiblock copolymer as

described below.

6.2.4.3. Synthesis of BisSF-BPS100 Multiblock Copolymers.

Oligomer (2) (Mn=3200 g/mol, 2.0244 g, 0.6135 mmol) was added to the solution

containing (1) and NMP (16 mL) was used to obtain a 14% w/v reaction solution. The bath

temperature was increased to 90 oC, and the reaction proceeded for 36 h. The resulting viscous

solution was precipitated into IPA (500 mL) to form fibrous strands. The product was filtered and

washed in deionized water at 60 oC for 12 h and acetone for 12 h. It was dried under vacuum at

110 oC for 24 h before casting into films.

174

6.2.5. Characterization of Copolymers.

1H and 19F NMR spectroscopy analyses were performed on a Varian INOVA 400

spectrometer operating at 400 and 376 MHz respectively. 13C NMR analyses were performed on

a Varian Unity 400 MHz spectrometer operating at 100 MHz. Spectra were obtained from a 10%

solution (w/v) solution in d6-DMSO at ambient temperature. Intrinsic viscosities of the

copolymers were determined using size exclusion chromatography (SEC). The experiments were

performed on a liquid chromatograph equipped with a Waters 1515 isocratic HPLC pump, Waters

Autosampler, Waters 2414 refractive index detector and Viscotek 270 dual detector. NMP

containing 0.05 M LiBr was used as the mobile phase. The column temperature was maintained

at 60 oC because of the viscous nature of NMP. Both the mobile phase and sample solution were

filtered before injection into the SEC system.

6.2.6. Membrane Preparation.

Membranes were cast from a 7% w/v solution of polymer in DMAc onto a clean glass

plate. Solvent was removed using an IR lamp. The film was held at 30-35 oC for 24 h and then

raised to 60 oC for an additional 24 h. It was dried under vacuum at 110 oC for 24 h. The film was

removed from the glass plate by submersion in water and acidified in boiling 0.5 M H2SO4 for 2

h, followed by 2 h in boiling deionized water.

6.2.7. Determination of Water Uptake and Dimensional Swelling.

The membrane water uptake was determined gravimetrically, using films with 100-150 mg

dry weight. Acidified membranes were equilibrated in water at room temperature for 24 h. Wet

membranes were removed from the water, blotted dry to remove excess water, and quickly

weighed. They were then dried at 110 oC under vacuum for 24 h and reweighed. Water uptake

175

was calculated according to Equation 1 where massdry and masswet refer to the mass of the dry and

wet membranes, respectively. An average of three samples was used for each measurement.

(1) 𝑊𝑎𝑡𝑒𝑟 𝑈𝑝𝑡𝑎𝑘𝑒 = 100 ∗𝑀𝑎𝑠𝑠𝑊𝑒𝑡−𝑀𝑎𝑠𝑠𝐷𝑟𝑦

𝑀𝑎𝑠𝑠𝐷𝑟𝑦

Percent swelling of the membranes was determined in the in-plane (x and y) and through-

plane (z) directions. Wet measurements were performed after equilibrating membranes in

deionized water for 24 h at room temperature. Membranes were then dried in a convection oven

at 80 oC for 2 h and measured again. Percent swelling was reported for three directions and

calculated according to Equation 2 where lengthwet,i and lengthdry,i refer to the length (where i

represents the x, y, or z direction) of the wet and dry membrane, respectively.

(2) 𝐷𝑖𝑚𝑒𝑛𝑡𝑖𝑜𝑛𝑎𝑙 𝑆𝑤𝑒𝑙𝑙𝑖𝑛𝑔 = 100 ∗𝐿𝑒𝑛𝑔𝑡ℎ𝑊𝑒𝑡,𝑖−𝐿𝑒𝑛𝑔𝑡ℎ𝐷𝑟𝑦,𝑖

𝐿𝑒𝑛𝑔𝑡ℎ𝐷𝑟𝑦,𝑖

6.2.8. Measurement of Proton Conductivity.

Proton conductivity at 30 oC in liquid water was determined in a window cell geometry

using a Solartron 1252 + 1287 Impedance/Gain-Phase Analyzer over the frequency range of 10

Hz to 1 MHz. In determining proton conductivity in liquid water, membranes were equilibrated at

30 oC in DI water for 24 h prior to testing.

Proton conductivity under partially hydrated conditions was measured at 80 oC.

Membranes were equilibrated at 80% RH for 8 h in a humidity-temperature oven (ESPEC, SH-

240). The conductivity measurements were made at varying RH beginning from high RH to low

RH. Membranes were allowed to equilibrate for 4 h prior to measuring conductivity at each RH.

Measurements were done at 95, 80, 70, 60, 50, 40 and 30%. Due to the changing membrane

thickness at different RH, film thickness was measured at 95, 80 and 30% RH using a micrometer.

176

The conductivity values at 70 and 60% RH were determined using the thickness measured at 80%

RH. The conductivity values at 50, 40 and 30% RH were determined using the thickness

measurement at 30% RH.

6.2.9. Tensile Testing.

Uniaxial load tests were performed using an Instron 5500R universal testing machine

equipped with a 200-lb load cell at room temperature and 44-54% RH. Crosshead displacement

speed was 5 mm/min and gauge length were set to 25 mm. The films were ~30 µm thick. A

dogbone die was used to punch specimens 50 mm long with a minimum width of 4 mm. Prior to

testing, specimens were dried under vacuum at 110 oC for at least 24 h and then equilibrated at

44% RH and 30 oC. All specimens were mounted in manually tightened grips. Approximate tensile

moduli for each specimen were calculated based on the stress and elongation values for the

specimen at 2% elongation.

6.2.10. Surface Morphology.

Tapping mode AFM was performed using a Digital Instruments MultiMode Scanning

Probe microscope with a NanoScope IVa controller. A Veeco silicon probe with an tip radius

<10 nm and a force constant K of 42 N/m was used to image samples. Samples were in their acid

form and equilibrated at 40% RH and 30 oC for at least 12 h prior to imaging.

6.2.11. Bulk Morphology.

For transmission electron microscopy imaging, acidified membranes were stained with a

lead acetate aqueous solution to enhance the electron density of the hydrophilic block and provide

contrast within the sample. Samples were embedded in epoxy Epo-fix (electron microscopy

177

services) and microtomed into approximately 70-nm thick sections with a diamond knife. Samples

were then imaged on a Philips EM 420 Transmission Electron Microscope using an accelerating

voltage of 100 kV.

6.3. Results and Discussion

6.3.1. Synthesis of Hydrophilic Oligomers.

Phenoxide-terminated, fully disulfonated poly(arylene ether sulfone) hydrophilic

oligomers (BPS100) were synthesized via nucleophilic aromatic substitution (Figure 6-1). A small

molar excess of BP relative to SDCDPS was used to control the molecular weight of the oligomers,

targeting number average molecular weights (Mn) of 3, 5, or 9 kg/mol. Proton NMR was used to

analyze the structures of the oligomers including both the biphenol endgroups and internal units.

Mn’s of the oligomers were calculated by ratioing the integrals of the endgroup peaks relative to

the internal units by assuming that all of the endgroups were phenolic (Figure 6-2). Protons from

the terminal BP groups were assigned to the peaks at 6.8, 7.05, 7.4, and 7.55 ppm, whereas protons

from the BP in the middle of the backbone resulted in peaks at 7.1 and 7.65 ppm. Theoretical and

experimental Mn values are summarized in Table 1, along with intrinsic viscosity (I.V.) measured

by SEC. An increase in I.V. was observed as Mn of the oligomers increased. A log-log plot of Mn

derived from the NMR spectra versus intrinsic viscosity (Figure 6-3) had a linear relationship,

suggesting successful control of molecular weight.

178

Figure 6-1. Phenoxide-Terminated BPS-100 with Controlled Molecular Weight

Figure 6-2. 1H NMR Spectrum of BPS-100 Oligomer

K2CO3

Toluene/DMAc

4 h @ 155 oC

96 h @ 190 oC

+OHOH Cl S Cl

O

OSO

3Na

NaO3S

OKO S O

O

OSO

3K

KO3S

OKn

K2CO3

Toluene/DMAc

4 h @ 155 oC

96 h @ 190 oC

+OHOH Cl S Cl

O

OSO

3Na

NaO3S

OKO S O

O

OSO

3K

KO3S

OKn

OKO S O

O

OSO

3K

KO3S

OKn

h b a

e

eg

g

ii d ch

f

f

OKO S O

O

OSO

3K

KO3S

OKn

h b a

e

eg

g

ii d ch

f

f

OKO S O

O

OSO

3K

KO3S

OKn

h b a

e

eg

g

ii d ch

f

f

4.03.916.

179

Table 6-1. Characterization of Hydrophilic Telechelic Oligomers

Molecular Weight

I.V. b

(dL/g) Target1 Experimentala

3000 2900 0.14

5000 4900 0.20

9000 9200 0.29

Figure 6-3. Log (Mn) vs log (I.V.) for the hydrophilic oligomers

y = 0.6082x - 2.9494

R2 = 0.9999

-0.9

-0.85

-0.8

-0.75

-0.7

-0.65

-0.6

-0.55

-0.5

3.4 3.6 3.8 4

Log (Mn)

Lo

g (

I.V

.)

Log(I.V)=0.61[Log(Mn)]-2.9y = 0.6082x - 2.9494

R2 = 0.9999

-0.9

-0.85

-0.8

-0.75

-0.7

-0.65

-0.6

-0.55

-0.5

3.4 3.6 3.8 4

Log (Mn)

Lo

g (

I.V

.)

Log(I.V)=0.61[Log(Mn)]-2.9

a) Calculated using 1H NMR

b) SEC of the oligomer in its salt form performed in NMP with 0.05 LiBr

180

6.3.2. Synthesis of BisSF-BPSH100 Segmented Copolymers.

The segmented copolymers were formed by reacting phenoxide-terminated hydrophilic

oligomers with DFBP and Bis-S monomers in a step growth polymerization as illustrated in Figure

6-4. Simultaneous formation of the hydrophobic segments and the block copolymer eliminated

the need to synthesize and isolate a separate hydrophobic block before coupling it to the

hydrophilic block. The stoichiometry was controlled such that the DFBP and Bis-S monomers

formed the hydrophobic segments of the copolymer, while also reacting with the phenoxide-

terminated hydrophilic oligomer. The molecular weights of the hydrophobic segments were

targeted to be 3, 5, or 9 kg/mol and copolymers with approximately equal hydrophobic and

hydrophilic block lengths were obtained. To achieve high molecular weight, it was important to

ensure that the overall ratio of phenoxide to para–fluorine endgroups was 1:1. An excess of

phenoxide groups was disadvantageous because once the para-fluorines were consumed, the

ortho-fluorines on the DFBP could react with the excess phenoxide groups, resulting in a

crosslinked network. However, if insufficient phenoxide groups were present, high molecular

weight copolymer was not achieved.

K2CO3

Cyclohexane/NMP

4 h @ 85 oC

Addition

of36-70 h @ 90 oC

(DFBP)

FFFF

F F F F

FF

S

O

O

OHOHOKO S O

O

OSO

3K

KO3S

OKn

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

K2CO3

Cyclohexane/NMP

4 h @ 85 oC

K2CO3

Cyclohexane/NMP

4 h @ 85 oC

Addition

of36-70 h @ 90 oC36-70 h @ 90 oC

(DFBP)

FFFF

F F F F

FF

S

O

O

OHOHOKO S O

O

OSO

3K

KO3S

OKn

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

181

Figure 6-4. BisSF-BPSH100 segmented copolymer

Representative 1H and 19F NMR spectra of a segmented copolymer are shown in Figure 6-

5. There were no peaks due to endgroups in either spectrum, which signifies successful segmented

copolymer formation. The peak at 7.3 ppm has been assigned to the protons of the BP in the

linking unit.

Figure 6-5. (a) 1H and (b) 19F NMR Spectra of a BisSF-BPS100 Segmented Copolymer

The highly reactive DFBP monomer allowed for mild reaction temperatures (90-110 oC)

to be used during synthesis, which eliminated randomization by ether-ether interchange.

Monomer sequencing is highly ordered in block copolymers. Therefore, most of the carbons have

only one possible chemical environment. Random copolymers develop a larger number of short

hi

(b)

(a)

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

a b c

e

d f g

h iFFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

a b c

e

d f g

h i

hi

(b)

(a)

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

a b c

e

d f g

h iFFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

a b c

e

d f g

h i

10.00 23.19

10.82 23.98 22.26

10.31 20.58

182

sequences during the copolymerization. This causes different chemical environments for similar

carbons, which results in splitting of the 13C NMR peaks. If substantial ether-ether interchange

occurred during a segmented copolymerization, a scrambling in the backbone would occur. This

would be evidenced by peak splitting similar to a random copolymer. The singlets in the 13C NMR

spectrum of a segmented copolymer were the same as for a multiblock copolymer with a similar

composition, both shown in Figure 6-6, which strongly supports the avoidance of ether-ether

interchange.

Figure 6-6. 13C NMR Spectra of BisSF-BPS100 Multiblock and Segmented Copolymers

6.3.3. Synthesis of BisSF-BPSH100 Multiblock Copolymer Controls.

BisSF-BPSH100 multiblock copolymer controls were synthesized as described in the

literature. Phenoxide-terminated, BPS100 oligomers were prepared using a slight molar excess of

BP to SDCDPS to control Mn. Hydrophobic oligomers (BisSF) were synthesized using a slight

molar excess of DFBP relative to Bis-S to achieve fluorine-terminated oligomers with controlled

Segmented 5K5K

Multiblock 5K5K

Segmented 5K5K

Multiblock 5K5K

183

Mn. Both hydrophilic and hydrophobic block lengths were targeted at 3, 5, or 9 kg/mol.

Experimental Mn’s were determined by endgroup analysis from 1H and 19F NMR for BPS100 and

BisSF oligomers, respectively, and agreed well with the target values (Table 6-2). BPS100 and

BisSF oligomers were coupled using a 1:1 molar ratio of the oligomers to form multiblock

copolymers with approximately equal hydrophilic and hydrophobic block lengths. The overall

synthetic procedure is depicted in Figure 6-7.

Table 6-2. Characterization of Hydrophilic and Hydrophobic Telechelic Oligomers for BisSF-BPSH100 Multiblock

Copolymers

Molecular Weight (Mn) (g/mol)

Target

BPSH100 BisSF

Experimentala Experimental

b

3000 3300 3200

5000 5000 6100

9000 9200 9300

184

Figure 6-7. Overview of Synthesis of BisSF-BPSH100 Multiblock Copolymer

6.3.4. Comparison of BisSF-BPSH100 Segmented and Multiblock Copolymer Properties.

The main objective of this research was to compare the properties of polymers produced

via a simplified synthetic technique relative to the traditional two-oligomer method. Select

properties of both systems are summarized in Table 6-3. Experimental IEC values were calculated

from 1H NMR by determining the weight ratios of the hydrophobic versus hydrophilic oligomers.

They were in good agreement with theoretical values, confirming that the hydrophobic segments

were systematically incorporated into the copolymer backbones. The I.V. data confirmed that high

K2CO3

Toluene/NMP

4 h @ 150 oC

36 h @ 190 oC

+ +

K2CO3

Cyclohexane/DMAc

5 h @ 110 oC

+

NMP

90-110 oC for 12-48 h

FFFF

F F F F

FF S

O

O

OHOHOHOH Cl S Cl

O

OSO

3Na

NaO3S

OKO S O

O

OSO

3K

KO3S

OKn

FFFF

F F F F

FFFF

F F F F

O S O

O

O

F m F

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

K2CO3

Toluene/NMP

4 h @ 150 oC

36 h @ 190 oC

+ +

K2CO3

Cyclohexane/DMAc

5 h @ 110 oC

+

NMP

90-110 oC for 12-48 h

FFFF

F F F F

FF S

O

O

OHOHOHOH Cl S Cl

O

OSO

3Na

NaO3S

OKO S O

O

OSO

3K

KO3S

OKn

FFFF

F F F F

FFFF

F F F F

O S O

O

O

F m F

FFFF

F F F F

FFFF

F F F F

O S O

O

O

OO S O

O

OSO

3K

KO3S

On

m

x

a) Calculated from 1H NMR

b) Calculated from 19F NMR

185

molecular weight polymers were achieved using both synthetic methods. When comparing

polymers with similar IEC values, water uptake increased with longer block lengths. Tensile

properties were also comparable for polymeric membranes prepared from both synthetic

techniques, as shown in Table 6-4.

Table 6-3. Characterization of Segmented and Multiblock BisSF-BPSH100 Copolymers

Polymer

IEC (meq/g) I.V.

(dL/g)

Conductivityc

(S/cm)

Water

Uptake

(%) Theoreticala Experimental

b

Segmented

3K-3K 1.6 1.8 0.63 0.10 62

5K-5K 1.6 1.5 0.50 0.11 51

9K-9K 1.7 1.5 0.82 0.15 74

Multiblock

3K-3K 1.8 1.8 1.07 0.13 73

5K-5K 1.6 1.6 0.89 0.13 78

9K-9K 1.7 1.6 0.84 0.13 101

a) Determined from reactant weights

b) Determined from 1H NMR

c) Determined in liquid water

186

Table 6-4. Comparison of Tensile Properties for Segmented and Multiblock Copolymers

Copolymer Modulus

(Mpa)

Tensile Strength

(Mpa)

Elongation

(%)

Max

Elongation

(%)

Segmented

3K-3K 1500±160 42±5 43±27 74

5K-5K 1470±120 39±2 16±5 21

9K-9K 1510±80 46±4 24±16 47

Multiblock

3K-3K 1380±200 42±3 43±15 55

5K-5K 1450±150 41±4 47±21 71

9K-9K 1390±200 43±3 22±4 26

Dimensional swelling was measured on both series of copolymers. The results for the

segmented and multiblock copolymers were also compared to a random copolymer (BPSH35)

(Figure 6-8). The segmented and multiblock copolymers exhibited anisotropic swelling in contrast

to the isotropic swelling of the random copolymer. Overall, swelling through the plane (z-

187

direction) increased with block length, while in-plane swelling (x- and y-directions) stayed the

same or decreased. This is likely indicative of the well-ordered morphology that develops as block

length increases. Membrane electrode failure, due to changes in humidity levels (swelling and

deswelling cycling), may be minimized with a reduction of in-plane swelling.

Figure 6-8. Comparison of Dimensional Swelling for Segmented, Multiblock, and Random Copolymers

The effect of block length on proton conductivity under partially hydrated conditions was

assessed for both systems. A plot of proton conductivity versus relative humidity is shown in

Figure 6-9. The segmented and multiblock copolymers had similar proton conductivities across

the entire RH range when comparing copolymers with equivalent block lengths. Conductivity also

increased over the entire RH range at greater block lengths. This indicates that increasing the

block length of the hydrophilic and hydrophobic segments improves the connectivity in the

hydrophilic channels, regardless of synthetic approach.

0

10

20

30

40

50

60

70

80

90

BPSH35 3K3K 3K3K 5K5K 5K5K 9K9K 9K9K

% S

well

ing

x y z

Segmented Segmented SegmentedMultiblock Multiblock Multiblock

z

X

y

Random0

10

20

30

40

50

60

70

80

90

BPSH35 3K3K 3K3K 5K5K 5K5K 9K9K 9K9K

% S

well

ing

x y z

Segmented Segmented SegmentedMultiblock Multiblock Multiblock

z

X

y

Random Segmented Segmented SegmentedMultiblock Multiblock Multiblock

z

X

yz

X

y

Random

188

Figure 6-9. Comparison of Proton Conductivity Under Partially Hydrated Conditions for Segmented and Multiblock Copolymers

TEM images shown in Figure 6-10 reflect both the development of long-range domain

order with an increase in block length from 5K-5K to 9K-9K, and also show similar trends for the

segmented and multiblock series. This is consistent with our observations of other multiblock

copolymer systems.25-30 The AFM images in Figure 6-11 similarly show the development of long-

range order with an increase in block length. This phase separated morphology correlates well with

the anisotropy of the dimensional water uptake data. Longer block lengths using either synthetic

method promote the formation of a lamellar morphology. As has been shown in other multiblock

systems, longer-range order correlates with reduced swelling in the X-Y directions (in-plane), and

increased Z-direction (through-plane) swelling. The hydrophobic lamellae may restrict in-plane

swelling in going from a dry to a fully humidified state. The hydrophobic domains may also reduce

water loss at reduced humidities, thus improving proton conductivity at low RH.

20 30 40 50 60 70 80 90 100

0.1

1

10

100

Pro

ton C

onductivity (

mS

/cm

)

Relative Humidity (%)

Segmented 3K3K

Multiblock 3K3K

Segmented 5K5K

Multiblock 5K5K

Segmented 9K9K

Multiblock 9K9K

Nafion 112

189

Figure 6-10. TEM images of (A) 5K-5K Multiblock, (B) 9K-9K Multiblock, (C) 5K-5K Segmented, and (D) 9K-9K

Segmented BisSF-BPSH100 copolymers. All images are taken at 96000 magnification.

190

Figure 6-11. AFM images of (A) 5K-5K Multiblock, (B) 9K-9K Multiblock, (C) 5K-5K Segmented, and (D) 9K-9K

Segmented BisSF-BPSH100 copolymers. All images are 1 micron by 1 micron.

6.4. Conclusions

Multiblock copolymers containing highly fluorinated hydrophobic blocks and 100%

disulfonated hydrophilic blocks have been synthesized using a simplified segmented approach.

The performance of these copolymers, particularly with respect to fuel cell operation, were in good

agreement with the properties of those synthesized using a more cumbersome oligomer-oligomer

approach. Increased conductivity over the entire RH range was evidence that a connected phase

morphology developed with longer block length. This morphology was confirmed by surface and

bulk investigations. The segmented method by all measures herein can be regarded as a potential

alternative for the production of phase-separated multiblock copolymers.

A) B)

C) D)

191

Acknowledgement. The authors would like to acknowledge the Department of Energy (DE-

FG36-06G016038) and NSF IGERT (DGE-0114346) for funding.

Reference:

1. Hickner, M. A.; Ghassemi, H.; Kim, Y. S.; Einsla, B. R.; McGrath, J. E., Alternative Polymer

Systems for Proton Exchange Membranes (PEMs). Chem. Rev. (Washington, DC, U. S.) 2004, 104 (10),

4587-4611.

2. Hamrock, S. J.; Herring, A. M.; Zawodzinski, T. A., Fuel cell chemistry and operation. J. Power

Sources 2007, 172 (1), 1.

3. Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.; Garland, N.; Myers, D.; Wilson,

M.; Garzon, F.; Wood, D.; Zelenay, P.; More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.; McGrath, J.

E.; Inaba, M.; Miyatake, K.; Hori, M.; Ota, K.; Ogumi, Z.; Miyata, S.; Nishikata, A.; Siroma, Z.;

Uchimoto, Y.; Yasuda, K.; Kimijima, K.-i.; Iwashita, N., Scientific Aspects of Polymer Electrolyte Fuel

Cell Durability and Degradation. Chem. Rev. (Washington, DC, U. S.) 2007, 107 (10), 3904-3951.

4. Lefebvre-Joud, F.; Mougin, J.; Antoni, L.; Bouyer, E.; Gebel, G.; Nony, F., Materials pertaining

to hydrogen production and conversion. Tech. Ing., Mater. Fonct. 2010, 8 (N 12), N1205/1-

N1205/21,Doc.N1205/1-Doc.N1205/4.

5. McVeigh, C., The hydrogen economy revisited: Solar Hydrogen: Moving Beyond Fossil Fuels

Joan M. Ogden and Robert H. Williams World Resources Institute, Washington, DC, USA, 1989,

US$10.00, 124 pp. Energy Policy 1990, 18 (2), 208-209.

6. White, R.; Yeh, S.; Goldstein, N. In Future water challenges of a hydrogen economy, American

Chemical Society: 2008; pp ENVR-189.

7. Mauritz, K. A.; Moore, R. B., State of understanding of Nafion. Chem. Rev. (Washington, DC, U.

S.) 2004, 104 (10), 4535-4585.

8. Hickner, M. A.; Pivovar, B. S., The chemical and structural nature of proton exchange membrane

fuel cell properties. Fuel Cells (Weinheim, Ger.) 2005, 5 (2), 213-229.

9. Iulianelli, A.; Basile, A., Sulfonated PEEK-based polymers in PEMFC and DMFC applications:

A review. International Journal of Hydrogen Energy 2012, 37 (20), 15241-15255.

10. Bae, B.; Miyatake, K.; Watanabe, M., Poly(arylene ether) block copolymer membranes:

synthesis, properties and durability. ECS Trans. 2009, 25 (1, Proton Exchange Membrane Fuel Cells 9),

415-422.

11. Im, S. J.; Patel, R.; Shin, S. J.; Kim, J. H.; Min, B. R., Sulfonated poly(arylene ether sulfone)

membranes based on biphenol for direct methanol fuel cells. Korean J. Chem. Eng. 2008, 25 (4), 732-

737.

12. Jung, M. S.; Kim, T.-H.; Yoon, Y. J.; Kang, C. G.; Yu, D. M.; Lee, J. Y.; Kim, H.-J.; Hong, Y.

T., Sulfonated poly(arylene sulfone) multiblock copolymers for proton exchange membrane fuel cells.


13. Matsumoto, K.; Ando, S.; Ueda, M., Synthesis of sulfonated poly(1,4-diphenoxybenzene) for

proton exchange membrane. Polym. J. (Tokyo, Jpn.) 2007, 39 (8), 882-887.

14. McGrath, J. E.; Roy, A.; Lee, H.-S.; Yu, X.; Badami, A.; Li, Y.; Wang, H., Multiblock

hydrophilic-hydrophobic proton exchange membranes for direct methanol based fuel cell. ECS Trans.

2007, 2 (24, Direct Methanol Fuel Cells), 55-62.

15. Miyatake, K.; Chikashige, Y.; Watanabe, M., Novel Sulfonated Poly(arylene ether): A Proton Conductive Polymer Electrolyte Designed for Fuel Cells. Macromolecules 2003, 36 (26), 9691-9693.

16. Nakabayashi, K.; Matsumoto, K.; Ueda, M., Synthesis and properties of sulfonated multiblock

copoly(ether sulfone)s by a chain extender. J. Polym. Sci., Part A: Polym. Chem. 2008, 46 (12), 3947-

3957.

192

17. Einsla, B. R.; Harris, L. A.; Hill, M. L.; McGrath, J. E., Direct copolymerization of wholly

aromatic and partially fluorinated disulfonated poly(arylene ether sulfone)s. Prepr. Symp. - Am. Chem.

Soc., Div. Fuel Chem. 2005, 50 (2), 571-572.

18. Roy, A.; Einsla, B. R.; Harrison, W. L.; McGrath, J. E., Synthesis and characterization of

hydroquinone based disulfonated poly (arylene ether sulfone)s via direct copolymerization. Prepr. Symp.

- Am. Chem. Soc., Div. Fuel Chem. 2004, 49 (2), 614-615.

19. Badami, A. S.; Roy, A.; Lee, H.-S.; Li, Y.; McGrath, J. E., Morphological investigations of

disulfonated poly(arylene ether sulfone)-b-naphthalene dianhydride-based polyimide multiblock

copolymers as potential high temperature proton exchange membranes. J. Membr. Sci. 2009, 328 (1+2),

156-164.

20. Lee, H.-S.; Roy, A.; Lane, O.; Lee, M.; McGrath, J. E., Synthesis and characterization of

multiblock copolymers based on hydrophilic disulfonated poly(arylene ether sulfone) and hydrophobic

partially fluorinated poly(arylene ether ketone) for fuel cell applications. J. Polym. Sci., Part A: Polym. Chem. 2010, 48 (1), 214-222.

21. Roy, A.; Lee, H.-S.; McGrath, J. E., Hydrophilic-hydrophobic multiblock copolymers based on

poly(arylene ether sulfone)s as novel proton exchange membranes - Part B. Polymer 2008, 49 (23), 5037-

5044. 22. Shin, D. W.; Lee, S. Y.; Lee, C. H.; Lee, K.-S.; Park, C. H.; McGrath, J. E.; Zhang, M.; Moore,

R. B.; Lingwood, M. D.; Madsen, L. A.; Kim, Y. T.; Hwang, I.; Lee, Y. M., Sulfonated Poly(arylene

sulfide sulfone nitrile) Multiblock Copolymers with Ordered Morphology for Proton Exchange

Membranes. Macromolecules (Washington, DC, U. S.) 2013, 46 (19), 7797-7804.

23. Yu, X.; Roy, A.; Dunn, S.; Badami, A. S.; Yang, J.; Good, A. S.; McGrath, J. E., Synthesis and

characterization of sulfonated-fluorinated, hydrophilic-hydrophobic multiblock copolymers for proton

exchange membranes. J. Polym. Sci., Part A: Polym. Chem. 2009, 47 (4), 1038-1051.

24. Sankir, M.; Bhanu, V. A.; Harrison, W. L.; Ghassemi, H.; Wiles, K. B.; Glass, T. E.; Brink, A. E.;

Brink, M. H.; McGrath, J. E., Synthesis and characterization of 3,3'-disulfonated-4,4'-dichlorodiphenyl

sulfone (SDCDPS) monomer for proton exchange membranes (PEM) in fuel cell applications. J. Appl.

Polym. Sci. 2006, 100 (6), 4595-4602. 25. Badami, A. S.; Lane, O.; Lee, H.-S.; Roy, A.; McGrath, J. E., Fundamental investigations of the

effect of the linkage group on the behavior of hydrophilic-hydrophobic poly(arylene ether sulfone)

multiblock copolymers for proton exchange membrane fuel cells. J. Membr. Sci. 2009, 333 (1+2), 1-11.

26. Badami, A. S.; Lee, H.-S.; Li, Y.; Roy, A.; Wang, H.; McGrath, J. E., Molecular weight effects

on poly(arylene ether sulfone)-based random and multiblock copolymers characteristics for fuel cells.

ACS Symp. Ser. 2010, 1040 (Fuel Cell Chemistry and Operation), 65-81.

27. Ghassemi, H.; Harrison, W.; Zawodzinski, T.; McGrath, J. E. In New multiblock copolymers containing hydrophilic-hydrophobic segments for proton exchange membranes, American Chemical

Society: 2004; pp POLY-420.

28. Lee, H.-S.; Lane, O. R.; McGrath, J. E. In Synthesis and characterization of multiblock copolymers with hydrophilic-hydrophobic sequences for proton exchange membranes, American

Chemical Society: 2008; pp FUEL-098.

29. Lee, H.-S.; Roy, A.; Lane, O.; Dunn, S.; McGrath, J. E., Hydrophilic-hydrophobic multiblock

copolymers based on poly(arylene ether sulfone) via low-temperature coupling reactions for proton

exchange membrane fuel cells. Polymer 2008, 49 (3), 715-723.

30. Lee, M.; Park, J. K.; Lee, H.-S.; Lane, O.; Moore, R. B.; McGrath, J. E.; Baird, D. G., Effects of

block length and solution-casting conditions on the final morphology and properties of disulfonated

poly(arylene ether sulfone) multiblock copolymer films for proton exchange membranes. Polymer 2009,

50 (25), 6129-6138.

193

7.0 Future Work

7.1 Post-sulfonated hydroquinone-containing poly(arylene ether sulfone)s The development of precisely tailored post-sulfonated hydroquinone-containing

copolymers has several interesting avenues for further investigation. The most obvious is the

fundamental work of synthesizing and sulfonating series with varying comonomer structures and

investigating the impact of the chemical composition on transport properties. The IECs

investigated in this research were fairly limited, and it would be informative to evaluate post-

sulfonated SHQS materials with a higher degree of hydrophilicity. This would be interesting at a

fundamental level to confirm if the calcium effect returns at higher sulfonate group

concentrations. Once a broad spectrum of IECs have been investigated for transport properties,

it might be informative to replace the bisphenyl sulfone with a more hydrophobic monomer. The

synthesis of a highly hydrophilic (60-80% hydroquinone-containing comonomers) oligomer,

which is then crosslinked in order to optimize both water flux and salt rejection, would be an

intuitive approach in the search for an improved membrane. It may also be informative in future

mixed-feed evaluations to compare a directly copolymerized (BPS) material that has been

exchanged with aluminum (Al3+), to gain a fundamental understanding of the impact of

counterion charge density and valence to limit or negate the screening effect of calcium.

However, the post-sulfonated poly(arylene ether sulfone) system may also provide a

novel approach for the synthesis of multiblock copolymers. The resistance to calcium

compromise of salt transport is an advantage unique to the application of reverse osmosis, but the

synthetic approach may afford an additional advantage for fuel cell or polymer membrane

electrolysis applications. While multiblock hydrophilic-hydrophobic poly(arylene ether sulfone)s

194

have been developed at Virginia Tech for a decade, the synthetic process remains lengthy and

delicate. Oligomer molecular weight, particularly of the hydrophilic block, is difficult to control

due to the poorly reactive sulfonated monomer and practical control over synthesis is very

limited.

In contrast, the synthesis of two nonsulfonated oligomers would be more rapid,

particularly if conducted at the high reaction temperatures of sulfolane, and the greater reactivity

should facilitate oligomer molecular weight control. A multiblock analogue of the series

described in Chapter 4, with a ‘pre-hydrophilic’ block composed of DCDPS and hydroquinone,

and a hydrophobic block of DCDPS and bisphenyl sulfone, would be comprised entirely of

commercially available monomer and very inexpensive to produce. After sulfonation and

recovery, the hydrophobic and hydrophilic oligomers could be coupled to produce a much more

feasible and affordable multiblock. This would allow for more rapid development and

optimization of multiblock architectures.

Alternately, it may also be possible to synthesize and couple the oligomers prior to

sulfonation of the final multiblock. While the hydrophobic component should not be soluble in

sulfuric acid by itself, the solubility of the reacting hydrophilic groups may be sufficient to keep

the multiblock dissolved.

One of the apparent hurdles to the development of post-sulfonated hydrophilic-

hydrophobic multiblock copolymers is the limited hydrophilicity of the monosulfonated unit.

Substituting the bisphenyl sulfone comonomer with for a deactivated dihalide with a single ring

would provide a final product with a similar ionic density. The spacing between the sulfonate

195

groups would not be a concern, and could possibly be advantageous in boosting connectivity

between hydrophilic domains of the resulting multiblock.

7.2 Structure-property-processing relationships in multiblock copolymers

The results published here on the impact of various casting conditions were remarkable in

how widely the final polymer performance varied as a consequence of casting solvent and

polymer concentration. Previously, few systematic studies were done on our hydrophilic-

hydrophobic blocks and their casting conditions. Future work in characterizing multiblock

materials should include a systematic evaluation of casting conditions and morphological

characterization. It may be desirable to revisit previously described systems that showed

competitive performance to commercial state-of-the-art membranes such as Nafion, in the event

that optimized casting conditions may provide superior performance. Due to its relative

simplicity and affordability, the potential post-sulfonated block copolymer described above

would be an ideal candidate.

Virginia Techii Synthesis, Characterization and Modification of Sulfonated Poly(arylene ether...

Documents

Transcript of Virginia Techii Synthesis, Characterization and Modification of Sulfonated Poly(arylene ether...